.claude/skills/content-researcher/references/dont-use-brand_guidelines.md

@@ -1,160 +0,0 @@
-# Brand Voice & Tone Guidelines
-
-Reference for maintaining consistent voice across all written content. These are defaults — override with client-specific guidelines when available.
-
----
-
-## Voice Archetypes
-
-Start with Expert but also work in Guide when appliciable.
-
-### Expert
-- **Sounds like:** A senior practitioner sharing hard-won knowledge.
-- **Characteristics:** Precise, evidence-backed, confident without arrogance. Cites data, references real-world experience, and isn't afraid to say "it depends."
-- **Typical vocabulary:** "In practice," "the tradeoff is," "based on our benchmarks," "here's why this matters."
-- **Risk to avoid:** Coming across as condescending or overly academic.
-- **Best for:** Technical audiences, B2B SaaS, engineering blogs, whitepapers.
-
-### Guide
-- **Sounds like:** A patient teacher walking you through something step by step.
-- **Characteristics:** Clear, encouraging, anticipates confusion. Breaks complex ideas into digestible pieces. Uses analogies.
-- **Typical vocabulary:** "Let's start with," "think of it like," "the key thing to remember," "don't worry if this seems complex."
-- **Risk to avoid:** Being patronizing or oversimplifying for an advanced audience.
-- **Best for:** Tutorials, onboarding content, documentation, beginner-to-intermediate audiences.
-
-
-
----
-
-## Core Writing Principles
-
-These apply regardless of archetype.
-
-### 1. Clarity First
-- If a sentence can be misread, rewrite it.
-- Use the simplest word that conveys the precise meaning. "Use" over "utilize." "Start" over "commence."
-- One idea per paragraph. One purpose per section.
-- Define jargon on first use, or skip it entirely.
-
-### 2. Customer-Centric
-- Frame everything from the reader's perspective, not the company's.
-- **Instead of:** "We built a new feature that enables real-time collaboration."
-- **Write:** "You can now edit documents with your team in real time."
-- Lead with the reader's problem or goal, not the product or solution.
-
-### 3. Active Voice
-- Active voice is the default. Passive voice is acceptable only when the actor is unknown or irrelevant.
-- **Active:** "The script generates a report every morning."
-- **Passive (acceptable):** "The logs are rotated every 24 hours." (The actor doesn't matter.)
-- **Passive (avoid):** "A decision was made to deprecate the endpoint." (Who decided?)
-
-### 4. Show, Don't Claim
-- Replace vague claims with specific evidence.
-- **Claim:** "Our platform is incredibly fast."
-- **Show:** "Queries return in under 50ms at the 99th percentile."
-- If you can't provide evidence, soften the language or cut the sentence.
-
----
-
-## Tone Attributes
-
-Tone shifts based on content type and audience. Use these spectrums to calibrate.
-
-### Formality Spectrum
-
-```
-Casual -------|-------|-------|-------|------- Formal
-  1       2       3       4       5
-```
-
-| Level | Description | Use When |
-|-------|-------------|----------|
-| 1 | Slang OK, sentence fragments, first person | Internal team comms, very informal blogs |
-| 2 | Conversational, contractions, direct address | Newsletters, community posts, most blog content |
-| 3 | Professional but approachable, minimal contractions | Product announcements, mid-funnel content |
-| 4 | Polished, structured, no contractions | Whitepapers, enterprise case studies, executive briefs |
-| 5 | Formal, third person, precise terminology | Legal, compliance, academic partnerships |
-
-**Default for most blog/article content: Level 2-3.**
-
-### Technical Depth Spectrum
-
-```
-General -------|-------|-------|-------|------- Deep Technical
-    1       2       3       4       5
-```
-
-| Level | Description | Use When |
-|-------|-------------|----------|
-| 1 | No jargon, analogy-heavy, conceptual | Non-technical stakeholders, general audience |
-| 2 | Light jargon (defined inline), practical focus | Business audience with some domain familiarity |
-| 3 | Industry-standard terminology, code snippets OK | Practitioners who do the work daily |
-| 4 | Assumes working knowledge, implementation details | Developers, engineers, technical decision-makers |
-| 5 | Deep internals, performance analysis, tradeoff math | Senior engineers, architects, researchers |
-
-**Default: Match the audience. When unsure, aim at what you think the audience can handle. We are mostly B2B.**
-
----
-
-## Language Preferences
-
-### Use Action Verbs
-Lead sentences — especially headings and CTAs — with strong verbs.
-
-| Weak | Strong |
-|------|--------|
-| There is a way to improve | Improve |
-| This section is a discussion of | This section covers |
-| You should consider using | Use |
-| It is important to note that | Note: |
-| We are going to walk through | Let's walk through |
-
-### Be Concrete and Specific
-Vague language erodes trust. Replace generalities with specifics.
-
-| Vague | Concrete |
-|-------|----------|
-| "significantly faster" | "3x faster" or "reduced from 12s to 2s" |
-| "a large number of users" | "over 40,000 monthly active users" |
-| "best-in-class" | describe the specific advantage |
-| "seamless integration" | "connects via a single API call" |
-| "in the near future" | "by Q2" or "in the next release" |
-
-### Avoid These Patterns
-- **Weasel words:** "very," "really," "extremely," "quite," "somewhat" — cut them or replace with data.
-- **Nominalizations:** "implementation" when you mean "implement," "utilization" when you mean "use."
-- **Hedge stacking:** "It might potentially be possible to perhaps consider..." — commit to a position or state the uncertainty once, clearly.
-- **Buzzword chains:** "AI-powered next-gen synergistic platform" — describe what it actually does.
-
----
-
-## Pre-Publication Checklist
-
-Run through this before publishing any piece of content.
-
-### Voice Consistency
-- [ ] Does the piece sound like one person wrote it, beginning to end?
-- [ ] Does it match the target voice archetype?
-- [ ] Are there jarring shifts in tone between sections?
-
-
-### Clarity
-- [ ] Can a reader in the target audience understand every sentence on the first read?
-- [ ] Is jargon defined or avoided?
-- [ ] Are all acronyms expanded on first use?
-- [ ] Do headings accurately describe the content beneath them?
-- [ ] Is the article scannable? (subheadings every 2-4 paragraphs, short paragraphs, lists where appropriate)
-
-### Value
-- [ ] Does the introduction make clear what the reader will gain?
-- [ ] Does every section earn its place? (Cut anything that doesn't serve the reader's goal.)
-- [ ] Are claims supported by evidence, examples, or data?
-- [ ] Is the advice actionable — can the reader do something with it today?
-- [ ] Does the conclusion provide a clear next step?
-
-### Formatting
-- [ ] Title includes the core keyword or topic and at least 2 closely related keyword's/topics.
-- [ ] Meta description summarizes the value proposition.
-- [ ] Code blocks, tables, and images have context (a sentence before them explaining what the reader is looking at).
-- [ ] Links use descriptive anchor text, not "click here."
-- [ ] No walls of text — maximum 5 sentences per paragraph for web content. Use a minimum of 2 sentences.

.claude/skills/content-researcher/scripts/cora_parser.py

@@ -59,13 +59,6 @@ OPTIMIZATION_RULES = {
     "exclude_measurement_entities": True,  # Ignore measurements (dimensions, tolerances) as entities
     "allow_organization_entities": True,   # Organizations like ISO, ANSI, etc. are OK
     "never_mention_competitors": True,     # Never mention competitors by name in content
-
-    # Entity correlation threshold
-    # Best of Both = lower of Spearman's or Pearson's correlation.
-    # Measures correlation to ranking position (1=top, 100=bottom), so negative = better ranking.
-    # Only include entities with Best of Both <= this value.
-    # Set to None to disable filtering.
-    "entity_correlation_threshold": -0.19,
 }
 
 
@@ -202,15 +195,6 @@ class CoraReport:
             if name.startswith("critical") or name.startswith("http"):
                 continue
 
-            correlation = _safe_float(row, col_map.get("Best of Both"))
-
-            # Filter by Best of Both correlation threshold.
-            # Lower (more negative) = stronger ranking signal (correlates with
-            # position 1 vs 100). Only keep entities at or below the threshold.
-            threshold = OPTIMIZATION_RULES.get("entity_correlation_threshold")
-            if threshold is not None and (correlation is None or correlation > threshold):
-                continue
-
             entity = {
                 "name": name,
                 "freebase_id": _safe_str(row, col_map.get("Freebase ID")),
@@ -219,7 +203,7 @@ class CoraReport:
                 "relevance": _safe_float(row, col_map.get("Relevance")),
                 "confidence": _safe_float(row, col_map.get("Confidence")),
                 "type": _safe_str(row, col_map.get("Type")),
-                "correlation": correlation,
+                "correlation": _safe_float(row, col_map.get("Best of Both")),
                 "current_count": _safe_int(row, site_col_idx),
                 "max_count": _safe_int(row, col_map.get("Max")),
                 "deficit": _safe_int(row, col_map.get("Deficit")),

.claude/skills/content-researcher/scripts/test_block_generator.py

@@ -433,7 +433,6 @@ def main():
 
     md_path = out_dir / "test_block.md"
     html_path = out_dir / "test_block.html"
-    txt_path = out_dir / "test_block.txt"
     stats_path = out_dir / "test_block_stats.json"
 
     md_content = format_markdown(result["sentences"])
@@ -441,7 +440,6 @@ def main():
 
     md_path.write_text(md_content, encoding="utf-8")
     html_path.write_text(html_content, encoding="utf-8")
-    txt_path.write_text(html_content, encoding="utf-8")
     stats_path.write_text(
         json.dumps(result["stats"], indent=2, default=str), encoding="utf-8"
     )
@@ -461,7 +459,6 @@ def main():
     print(f"\nFiles written:")
     print(f"  {md_path}")
     print(f"  {html_path}")
-    print(f"  {txt_path}")
     print(f"  {stats_path}")
 
 

.claude/skills/content-researcher/skill.md

@@ -492,7 +492,6 @@ These override any data from the Cora report:
 | Competitor names | NEVER use competitor company names as entities or LSI keywords. Do not mention competitors by name in content. |
 | Measurement entities | Ignore measurements (dimensions, tolerances, etc.) as entities — skip these in entity optimization |
 | Organization entities | Organizations like ISO, ANSI, ASTM are fine — keep these as entities |
-| Entity correlation filter | Only entities with Best of Both <= -0.19 are included. Best of Both is the lower of Spearman's or Pearson's correlation to ranking position (1=top, 100=bottom), so more negative = stronger ranking signal. This filter is applied in `cora_parser.py` and affects all downstream consumers. To disable, set `entity_correlation_threshold` to `None` in `OPTIMIZATION_RULES`. Added 2026-03-20 — revert if entity coverage feels too thin. |
 
 ---
 

cheddahbot/__main__.py

@@ -17,13 +17,13 @@ logging.basicConfig(
     datefmt="%H:%M:%S",
 )
 
-# All levels to rotating log file (DEBUG+)
+# Warnings and errors to rotating log file
 _log_dir = Path(__file__).resolve().parent.parent / "logs"
 _log_dir.mkdir(exist_ok=True)
 _file_handler = RotatingFileHandler(
     _log_dir / "cheddahbot.log", maxBytes=5 * 1024 * 1024, backupCount=5
 )
-_file_handler.setLevel(logging.DEBUG)
+_file_handler.setLevel(logging.WARNING)
 _file_handler.setFormatter(
     logging.Formatter("%(asctime)s [%(name)s] %(levelname)s: %(message)s")
 )
@@ -181,14 +181,22 @@ def main():
         log.info("Starting scheduler...")
         scheduler = Scheduler(config, db, default_agent, notification_bus=notification_bus)
         scheduler.start()
-        # Inject scheduler into tool context so get_active_tasks can read it
-        if tools:
-            tools.scheduler = scheduler
     except Exception as e:
         log.warning("Scheduler not available: %s", e)
 
+    log.info("Launching Gradio UI on %s:%s...", config.host, config.port)
+    blocks = create_ui(
+        registry, config, default_llm, notification_bus=notification_bus, scheduler=scheduler
+    )
+
+    # Build a parent FastAPI app so we can mount the dashboard alongside Gradio.
+    # Inserting routes into blocks.app before launch() doesn't work because
+    # launch()/mount_gradio_app() replaces the internal App instance.
+    import gradio as gr
     import uvicorn
     from fastapi import FastAPI
+    from fastapi.responses import RedirectResponse
+    from starlette.staticfiles import StaticFiles
 
     fastapi_app = FastAPI()
 
@@ -199,33 +207,24 @@ def main():
     fastapi_app.include_router(api_router)
     log.info("API router mounted at /api/")
 
-    # Mount new HTMX web UI (chat at /, dashboard at /dashboard)
+    # Mount the dashboard as static files (must come before Gradio's catch-all)
-    from .web import mount_web_app
+    dashboard_dir = Path(__file__).resolve().parent.parent / "dashboard"
+    if dashboard_dir.is_dir():
+        # Redirect /dashboard (no trailing slash) → /dashboard/
+        @fastapi_app.get("/dashboard")
+        async def _dashboard_redirect():
+            return RedirectResponse(url="/dashboard/")
 
-    mount_web_app(
+        fastapi_app.mount(
-        fastapi_app,
+            "/dashboard",
-        registry,
+            StaticFiles(directory=str(dashboard_dir), html=True),
-        config,
+            name="dashboard",
-        default_llm,
-        notification_bus=notification_bus,
-        scheduler=scheduler,
-        db=db,
-    )
-
-    # Mount Gradio at /old for transition period
-    try:
-        import gradio as gr
-
-        log.info("Mounting Gradio UI at /old...")
-        blocks = create_ui(
-            registry, config, default_llm, notification_bus=notification_bus, scheduler=scheduler
         )
-        gr.mount_gradio_app(fastapi_app, blocks, path="/old", pwa=False, show_error=True)
+        log.info("Dashboard mounted at /dashboard/ (serving %s)", dashboard_dir)
-        log.info("Gradio UI available at /old")
+
-    except Exception as e:
+    # Mount Gradio at the root
-        log.warning("Gradio UI not available: %s", e)
+    gr.mount_gradio_app(fastapi_app, blocks, path="/", pwa=True, show_error=True)
 
-    log.info("Launching web UI on %s:%s...", config.host, config.port)
     uvicorn.run(fastapi_app, host=config.host, port=config.port)
 
 

cheddahbot/agent.py

@@ -382,10 +382,7 @@ class Agent:
             skip_permissions: If True, run CLI with --dangerously-skip-permissions.
         """
         log.info("Execution brain task: %s", prompt[:100])
-        kwargs: dict = {
+        kwargs: dict = {"system_prompt": system_context}
-            "system_prompt": system_context,
-            "timeout": self.config.timeouts.execution_brain,
-        }
         if tools:
             kwargs["tools"] = tools
         if model:
@@ -397,7 +394,7 @@ class Agent:
         # Log to daily memory
         if self._memory:
             try:
-                self._memory.log_daily(f"[Execution] {prompt[:200]}\n-> {result[:500]}")
+                self._memory.log_daily(f"[Execution] {prompt[:200]}\n→ {result[:500]}")
             except Exception as e:
                 log.warning("Failed to log execution to memory: %s", e)
 

cheddahbot/config.py

@@ -98,12 +98,6 @@ class ApiBudgetConfig:
     alert_threshold: float = 0.8  # alert at 80% of limit
 
 
-@dataclass
-class TimeoutConfig:
-    execution_brain: int = 2700  # 45 minutes
-    blm: int = 1800  # 30 minutes
-
-
 @dataclass
 class ContentConfig:
     cora_inbox: str = ""  # e.g. "Z:/content-cora-inbox"
@@ -163,7 +157,6 @@ class Config:
     autocora: AutoCoraConfig = field(default_factory=AutoCoraConfig)
     api_budget: ApiBudgetConfig = field(default_factory=ApiBudgetConfig)
     content: ContentConfig = field(default_factory=ContentConfig)
-    timeouts: TimeoutConfig = field(default_factory=TimeoutConfig)
     ntfy: NtfyConfig = field(default_factory=NtfyConfig)
     agents: list[AgentConfig] = field(default_factory=lambda: [AgentConfig()])
 
@@ -228,10 +221,6 @@ def load_config() -> Config:
             for k, v in data["content"].items():
                 if hasattr(cfg.content, k):
                     setattr(cfg.content, k, v)
-        if "timeouts" in data and isinstance(data["timeouts"], dict):
-            for k, v in data["timeouts"].items():
-                if hasattr(cfg.timeouts, k):
-                    setattr(cfg.timeouts, k, int(v))
 
         # ntfy push notifications
         if "ntfy" in data and isinstance(data["ntfy"], dict):
@@ -299,12 +288,6 @@ def load_config() -> Config:
     if blm_dir := os.getenv("BLM_DIR"):
         cfg.link_building.blm_dir = blm_dir
 
-    # Timeout env var overrides (seconds)
-    if t := os.getenv("CHEDDAH_TIMEOUT_EXECUTION_BRAIN"):
-        cfg.timeouts.execution_brain = int(t)
-    if t := os.getenv("CHEDDAH_TIMEOUT_BLM"):
-        cfg.timeouts.blm = int(t)
-
     # Ensure data directories exist
     cfg.data_dir.mkdir(parents=True, exist_ok=True)
     (cfg.data_dir / "uploads").mkdir(exist_ok=True)

cheddahbot/llm.py

@@ -157,7 +157,6 @@ class LLMAdapter:
         tools: str = "Bash,Read,Edit,Write,Glob,Grep",
         model: str | None = None,
         skip_permissions: bool = False,
-        timeout: int = 2700,
     ) -> str:
         """Execution brain: calls Claude Code CLI with full tool access.
 
@@ -168,7 +167,6 @@ class LLMAdapter:
             tools: Comma-separated Claude Code tool names (default: standard set).
             model: Override the CLI model (e.g. "claude-sonnet-4.5").
             skip_permissions: If True, append --dangerously-skip-permissions to
-            timeout: Max seconds to wait for CLI completion (default: 2700 / 45 min).
                 the CLI invocation (used for automated pipelines).
         """
         claude_bin = shutil.which("claude")
@@ -220,11 +218,10 @@ class LLMAdapter:
             )
 
         try:
-            stdout, stderr = proc.communicate(input=prompt, timeout=timeout)
+            stdout, stderr = proc.communicate(input=prompt, timeout=900)
         except subprocess.TimeoutExpired:
             proc.kill()
-            minutes = timeout // 60
+            return "Error: Claude Code execution timed out after 15 minutes."
-            return f"Error: Claude Code execution timed out after {minutes} minutes."
 
         if proc.returncode != 0:
             return f"Execution error: {stderr or 'unknown error'}"

cheddahbot/scheduler.py

@@ -8,8 +8,6 @@ import logging
 import re
 import shutil
 import threading
-
-import httpx
 from datetime import UTC, datetime
 from pathlib import Path
 from typing import TYPE_CHECKING
@@ -25,26 +23,8 @@ if TYPE_CHECKING:
 
 log = logging.getLogger(__name__)
 
-# Dedicated logger for "tool returned error but likely handled it" cases.
-# Writes to logs/pipeline_errors.log for manual review.
-_pipeline_err_log = logging.getLogger("cheddahbot.pipeline_errors")
-_pipeline_err_log.propagate = False
-_pe_dir = Path(__file__).resolve().parent.parent / "logs"
-_pe_dir.mkdir(exist_ok=True)
-_pe_handler = logging.FileHandler(_pe_dir / "pipeline_errors.log", encoding="utf-8")
-_pe_handler.setFormatter(
-    logging.Formatter("%(asctime)s | %(message)s")
-)
-_pipeline_err_log.addHandler(_pe_handler)
-_pipeline_err_log.setLevel(logging.INFO)
-
 HEARTBEAT_OK = "HEARTBEAT_OK"
 
-# Only tasks in these statuses are eligible for xlsx ->ClickUp matching.
-# "to do" is excluded to prevent accidental matches and AutoCora race conditions.
-# To force-reuse an xlsx for a "to do" task, set status to "running cora" first.
-_CORA_ELIGIBLE_STATUSES = frozenset({"running cora", "error"})
-
 
 class Scheduler:
     # Tasks due within this window are eligible for execution
@@ -87,62 +67,6 @@ class Scheduler:
             "cora_distribute": None,
             "briefing": None,
         }
-        self._active_executions: dict[str, dict] = {}
-        self._active_lock = threading.Lock()
-        self._plural_cache: dict[tuple[str, str], bool] = {}
-
-    def _llm_plural_check(self, a: str, b: str) -> bool:
-        """Ask the chat brain if two keywords are the same aside from plural form.
-
-        Uses OpenRouter with the configured CHEDDAH_CHAT_MODEL. Results are
-        cached for the session to avoid repeat calls.
-        """
-        key = (a, b) if a <= b else (b, a)
-        if key in self._plural_cache:
-            return self._plural_cache[key]
-
-        api_key = self.config.openrouter_api_key
-        model = self.config.chat_model
-        if not api_key:
-            log.warning("LLM plural check: no OpenRouter API key, returning False")
-            return False
-
-        try:
-            resp = httpx.post(
-                "https://openrouter.ai/api/v1/chat/completions",
-                headers={"Authorization": f"Bearer {api_key}"},
-                json={
-                    "model": model,
-                    "max_tokens": 5,
-                    "messages": [
-                        {
-                            "role": "system",
-                            "content": (
-                                "You compare SEO keywords. Reply with ONLY 'YES' or 'NO'. "
-                                "Answer YES only if the two keywords are identical except for "
-                                "singular vs plural word forms (e.g. 'shaft' vs 'shafts', "
-                                "'company' vs 'companies'). Answer NO if they differ in any "
-                                "other way (extra words, different words, different meaning)."
-                            ),
-                        },
-                        {
-                            "role": "user",
-                            "content": f'Keyword A: "{a}"\nKeyword B: "{b}"',
-                        },
-                    ],
-                },
-                timeout=15,
-            )
-            resp.raise_for_status()
-            answer = (resp.json()["choices"][0]["message"]["content"] or "").strip()
-            result = "YES" in answer.upper()
-            log.debug("LLM plural check: '%s' vs '%s' ->%s (%s)", a, b, result, answer)
-        except Exception as e:
-            log.warning("LLM plural check failed for '%s' vs '%s': %s", a, b, e)
-            result = False
-
-        self._plural_cache[key] = result
-        return result
 
     def start(self):
         """Start the scheduler, heartbeat, and ClickUp threads."""
@@ -285,26 +209,6 @@ class Scheduler:
         """Return last_run timestamps for all loops (in-memory)."""
         return dict(self._loop_timestamps)
 
-    def _register_execution(self, task_id: str, name: str, tool_name: str) -> None:
-        """Register a task as actively executing."""
-        with self._active_lock:
-            self._active_executions[task_id] = {
-                "name": name,
-                "tool": tool_name,
-                "started_at": datetime.now(UTC),
-                "thread": threading.current_thread().name,
-            }
-
-    def _unregister_execution(self, task_id: str) -> None:
-        """Remove a task from the active executions registry."""
-        with self._active_lock:
-            self._active_executions.pop(task_id, None)
-
-    def get_active_executions(self) -> dict[str, dict]:
-        """Return a snapshot of currently executing tasks."""
-        with self._active_lock:
-            return dict(self._active_executions)
-
     # ── Scheduled Tasks ──
 
     def _poll_loop(self):
@@ -488,7 +392,7 @@ class Scheduler:
                 status_triggers = mapping.get("auto_execute_on_status", [])
                 if task.status.lower() not in [s.lower() for s in status_triggers]:
                     hint = mapping.get("trigger_hint", "manual trigger only")
-                    log.debug(
+                    log.info(
                         "Skipping task '%s' (type=%s): auto_execute is false (%s)",
                         task.name,
                         task.task_type,
@@ -542,10 +446,9 @@ class Scheduler:
         # Move to "automation underway" on ClickUp immediately
         client.update_task_status(task_id, self.config.clickup.automation_status)
 
-        log.info("Executing ClickUp task: %s ->%s", task.name, tool_name)
+        log.info("Executing ClickUp task: %s → %s", task.name, tool_name)
-        self._notify(f"Executing ClickUp task: **{task.name}** ->Skill: `{tool_name}`")
+        self._notify(f"Executing ClickUp task: **{task.name}** → Skill: `{tool_name}`")
 
-        self._register_execution(task_id, task.name, tool_name)
         try:
             # args already built during validation above
             args["clickup_task_id"] = task_id
@@ -571,12 +474,12 @@ class Scheduler:
             if result.startswith("Skipped:") or result.startswith("Error:"):
                 client.add_comment(
                     task_id,
-                    f"[WARNING]CheddahBot could not execute this task.\n\n{result[:2000]}",
+                    f"⚠️ CheddahBot could not execute this task.\n\n{result[:2000]}",
                 )
                 client.update_task_status(task_id, self.config.clickup.error_status)
 
                 self._notify(f"ClickUp task skipped: **{task.name}**\nReason: {result[:200]}")
-                log.debug("ClickUp task skipped: %s — %s", task.name, result[:200])
+                log.info("ClickUp task skipped: %s — %s", task.name, result[:200])
                 return
 
             # Tool handled its own ClickUp sync — just log success
@@ -585,7 +488,7 @@ class Scheduler:
 
         except Exception as e:
             client.add_comment(
-                task_id, f"[FAILED]CheddahBot failed to complete this task.\n\nError: {str(e)[:2000]}"
+                task_id, f"❌ CheddahBot failed to complete this task.\n\nError: {str(e)[:2000]}"
             )
             client.update_task_status(task_id, self.config.clickup.error_status)
 
@@ -594,8 +497,6 @@ class Scheduler:
                 f"Skill: `{tool_name}` | Error: {str(e)[:200]}"
             )
             log.error("ClickUp task failed: %s — %s", task.name, e)
-        finally:
-            self._unregister_execution(task_id)
 
     def _recover_stale_tasks(self):
         """Reset tasks stuck in 'automation underway' for too long.
@@ -641,7 +542,7 @@ class Scheduler:
                 client.update_task_status(task.id, reset_status)
                 client.add_comment(
                     task.id,
-                    f"[WARNING]CheddahBot auto-recovered this task. It was stuck in "
+                    f"⚠️ CheddahBot auto-recovered this task. It was stuck in "
                     f"'{automation_status}' for {age_ms / 3_600_000:.1f} hours. "
                     f"Reset to '{reset_status}' for retry.",
                 )
@@ -758,7 +659,7 @@ class Scheduler:
                     category="autocora",
                 )
 
-            log.debug("AutoCora result for '%s': %s", keyword, status)
+            log.info("AutoCora result for '%s': %s", keyword, status)
 
             # Move result file to processed/
             processed_dir.mkdir(exist_ok=True)
@@ -821,7 +722,7 @@ class Scheduler:
         """Try to match a watched .xlsx file to a ClickUp task and run the pipeline."""
         filename = xlsx_path.name
         # Normalize filename stem for matching
-        # e.g., "precision-cnc-machining" ->"precision cnc machining"
+        # e.g., "precision-cnc-machining" → "precision cnc machining"
         stem = xlsx_path.stem.lower().replace("-", " ").replace("_", " ")
         stem = re.sub(r"\s+", " ", stem).strip()
 
@@ -858,11 +759,6 @@ class Scheduler:
         money_site_url = matched_task.custom_fields.get("IMSURL", "") or ""
         if not money_site_url:
             log.warning("Task %s (%s) missing IMSURL — skipping", task_id, matched_task.name)
-            client.add_comment(
-                task_id,
-                "[FAILED]Link building skipped — IMSURL field is empty. "
-                "Set the IMSURL field in ClickUp so the pipeline knows where to build links.",
-            )
             client.update_task_status(task_id, self.config.clickup.error_status)
             self._notify(
                 f"Folder watcher: **{filename}** matched task **{matched_task.name}** "
@@ -887,7 +783,6 @@ class Scheduler:
             with contextlib.suppress(ValueError, TypeError):
                 args["branded_plus_ratio"] = float(bp_raw)
 
-        self._register_execution(task_id, matched_task.name, "run_cora_backlinks")
         try:
             # Execute via tool registry
             if hasattr(self.agent, "_tools") and self.agent._tools:
@@ -897,10 +792,6 @@ class Scheduler:
 
             if "Error" in result and "## Step" not in result:
                 # Pipeline failed — tool handles its own ClickUp error status
-                _pipeline_err_log.info(
-                    "LINKBUILDING | task=%s | file=%s | result=%s",
-                    task_id, filename, result[:500],
-                )
                 self._notify(
                     f"Folder watcher: pipeline **failed** for **{filename}**.\n"
                     f"Error: {result[:200]}",
@@ -925,13 +816,7 @@ class Scheduler:
 
         except Exception as e:
             log.error("Folder watcher pipeline error for %s: %s", filename, e)
-            client.add_comment(
-                task_id,
-                f"[FAILED]Link building pipeline crashed.\n\nError: {str(e)[:2000]}",
-            )
             client.update_task_status(task_id, self.config.clickup.error_status)
-        finally:
-            self._unregister_execution(task_id)
 
     def _match_xlsx_to_clickup(self, normalized_stem: str):
         """Find a ClickUp Link Building task whose Keyword matches the file stem.
@@ -946,14 +831,12 @@ class Scheduler:
             return None
 
         try:
-            tasks = client.get_tasks_from_overall_lists(space_id, statuses=list(_CORA_ELIGIBLE_STATUSES))
+            tasks = client.get_tasks_from_overall_lists(space_id)
         except Exception as e:
             log.warning("ClickUp query failed in _match_xlsx_to_clickup: %s", e)
             return None
 
         for task in tasks:
-            if task.status not in _CORA_ELIGIBLE_STATUSES:
-                continue
             if task.task_type != "Link Building":
                 continue
 
@@ -966,7 +849,7 @@ class Scheduler:
                 continue
 
             keyword_norm = _normalize_for_match(str(keyword))
-            if _fuzzy_keyword_match(normalized_stem, keyword_norm, self._llm_plural_check):
+            if _fuzzy_keyword_match(normalized_stem, keyword_norm):
                 return task
 
         return None
@@ -1064,7 +947,6 @@ class Scheduler:
             "clickup_task_id": task_id,
         }
 
-        self._register_execution(task_id, matched_task.name, "create_content")
         try:
             if hasattr(self.agent, "_tools") and self.agent._tools:
                 result = self.agent._tools.execute("create_content", args)
@@ -1072,10 +954,6 @@ class Scheduler:
                 result = "Error: tool registry not available"
 
             if result.startswith("Error:"):
-                _pipeline_err_log.info(
-                    "CONTENT | task=%s | file=%s | result=%s",
-                    task_id, filename, result[:500],
-                )
                 self._notify(
                     f"Content watcher: pipeline **failed** for **{filename}**.\n"
                     f"Error: {result[:200]}",
@@ -1100,13 +978,6 @@ class Scheduler:
 
         except Exception as e:
             log.error("Content watcher pipeline error for %s: %s", filename, e)
-            client.add_comment(
-                task_id,
-                f"[FAILED]Content pipeline crashed.\n\nError: {str(e)[:2000]}",
-            )
-            client.update_task_status(task_id, self.config.clickup.error_status)
-        finally:
-            self._unregister_execution(task_id)
 
     def _match_xlsx_to_content_task(self, normalized_stem: str):
         """Find a ClickUp content task whose Keyword matches the file stem.
@@ -1122,15 +993,13 @@ class Scheduler:
             return None
 
         try:
-            tasks = client.get_tasks_from_overall_lists(space_id, statuses=list(_CORA_ELIGIBLE_STATUSES))
+            tasks = client.get_tasks_from_overall_lists(space_id)
         except Exception as e:
             log.warning("ClickUp query failed in _match_xlsx_to_content_task: %s", e)
             return None
 
         content_types = ("Content Creation", "On Page Optimization")
         for task in tasks:
-            if task.status not in _CORA_ELIGIBLE_STATUSES:
-                continue
             if task.task_type not in content_types:
                 continue
 
@@ -1139,7 +1008,7 @@ class Scheduler:
                 continue
 
             keyword_norm = _normalize_for_match(str(keyword))
-            if _fuzzy_keyword_match(normalized_stem, keyword_norm, self._llm_plural_check):
+            if _fuzzy_keyword_match(normalized_stem, keyword_norm):
                 return task
 
         return None
@@ -1207,7 +1076,7 @@ class Scheduler:
             return
 
         try:
-            tasks = client.get_tasks_from_overall_lists(space_id, statuses=list(_CORA_ELIGIBLE_STATUSES))
+            tasks = client.get_tasks_from_overall_lists(space_id)
         except Exception as e:
             log.warning("ClickUp query failed in _distribute_cora_file: %s", e)
             return
@@ -1216,22 +1085,17 @@ class Scheduler:
         has_lb = False
         has_content = False
         matched_names = []
-        matched_error_tasks = []
 
         for task in tasks:
-            if task.status not in _CORA_ELIGIBLE_STATUSES:
-                continue
             keyword = task.custom_fields.get("Keyword", "")
             if not keyword:
                 continue
 
             keyword_norm = _normalize_for_match(str(keyword))
-            if not _fuzzy_keyword_match(stem, keyword_norm, self._llm_plural_check):
+            if not _fuzzy_keyword_match(stem, keyword_norm):
                 continue
 
             matched_names.append(task.name)
-            if task.status == self.config.clickup.error_status:
-                matched_error_tasks.append(task)
             if task.task_type == "Link Building":
                 has_lb = True
             elif task.task_type in ("Content Creation", "On Page Optimization"):
@@ -1268,19 +1132,6 @@ class Scheduler:
             )
             return
 
-        # Reset any matched tasks that were in "error" back to "running cora"
-        # so the pipeline picks them up again.
-        for task in matched_error_tasks:
-            try:
-                client.update_task_status(task.id, "running cora")
-                client.add_comment(
-                    task.id,
-                    f"New Cora XLSX distributed — resetting from error to running cora.",
-                )
-                log.info("Distributor: reset task %s (%s) from error ->running cora", task.id, task.name)
-            except Exception as e:
-                log.warning("Distributor: failed to reset task %s: %s", task.id, e)
-
         # Move original to processed/
         processed_dir = xlsx_path.parent / "processed"
         processed_dir.mkdir(exist_ok=True)
@@ -1289,7 +1140,7 @@ class Scheduler:
         except OSError as e:
             log.warning("Could not move %s to processed: %s", filename, e)
 
-        log.info("Cora distributor: %s ->%s", filename, ", ".join(copied_to))
+        log.info("Cora distributor: %s → %s", filename, ", ".join(copied_to))
         self._notify(
             f"Cora distributor: **{filename}** copied to {', '.join(copied_to)}.\n"
             f"Matched tasks: {', '.join(matched_names)}",
@@ -1352,7 +1203,7 @@ class Scheduler:
                 if not keyword:
                     continue
                 keyword_norm = _normalize_for_match(str(keyword))
-                if _fuzzy_keyword_match(normalized_stem, keyword_norm, self._llm_plural_check):
+                if _fuzzy_keyword_match(normalized_stem, keyword_norm):
                     task_ids.append(task.id)
 
         # Post comments
@@ -1505,7 +1356,7 @@ class Scheduler:
     def _check_cora_file_status(self, cora_tasks) -> dict[str, str]:
         """For each 'running cora' task, check where its xlsx sits on the network.
 
-        Returns a dict of task_id ->human-readable status note.
+        Returns a dict of task_id → human-readable status note.
         """
         from .tools.linkbuilding import _fuzzy_keyword_match, _normalize_for_match
 
@@ -1519,7 +1370,7 @@ class Scheduler:
             "content_processed": Path(self.config.content.cora_inbox) / "processed",
         }
 
-        # Build a map: normalized_stem ->set of folder keys
+        # Build a map: normalized_stem → set of folder keys
         file_locations: dict[str, set[str]] = {}
         for folder_key, folder_path in folders.items():
             if not folder_path.exists():
@@ -1540,7 +1391,7 @@ class Scheduler:
             # Find which folders have a matching file
             matched_folders: set[str] = set()
             for stem, locs in file_locations.items():
-                if _fuzzy_keyword_match(keyword_norm, stem, self._llm_plural_check):
+                if _fuzzy_keyword_match(keyword_norm, stem):
                     matched_folders.update(locs)
 
             if not matched_folders:

cheddahbot/static/app.css

@@ -1,619 +0,0 @@
-/* CheddahBot Dark Theme */
-
-:root {
-    --bg-primary: #0d1117;
-    --bg-surface: #161b22;
-    --bg-surface-hover: #1c2129;
-    --bg-input: #0d1117;
-    --text-primary: #e6edf3;
-    --text-secondary: #8b949e;
-    --text-muted: #484f58;
-    --accent: #2dd4bf;
-    --accent-dim: #134e4a;
-    --border: #30363d;
-    --success: #3fb950;
-    --error: #f85149;
-    --warning: #d29922;
-    --font-sans: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif;
-    --font-mono: 'JetBrains Mono', 'Fira Code', 'Cascadia Code', monospace;
-    --radius: 8px;
-    --sidebar-width: 280px;
-}
-
-* { margin: 0; padding: 0; box-sizing: border-box; }
-
-html, body {
-    height: 100%;
-    font-family: var(--font-sans);
-    font-size: 15px;
-    line-height: 1.5;
-    color: var(--text-primary);
-    background: var(--bg-primary);
-    overflow: hidden;
-}
-
-/* Top Navigation */
-.top-nav {
-    display: flex;
-    align-items: center;
-    gap: 24px;
-    padding: 0 20px;
-    height: 48px;
-    background: var(--bg-surface);
-    border-bottom: 1px solid var(--border);
-    flex-shrink: 0;
-}
-
-.nav-brand {
-    font-weight: 700;
-    font-size: 1.1em;
-    color: var(--accent);
-}
-
-.nav-links { display: flex; gap: 4px; }
-
-.nav-link {
-    color: var(--text-secondary);
-    text-decoration: none;
-    padding: 6px 14px;
-    border-radius: var(--radius);
-    font-size: 0.9em;
-    transition: background 0.15s, color 0.15s;
-}
-.nav-link:hover { background: var(--bg-surface-hover); color: var(--text-primary); }
-.nav-link.active { color: var(--accent); background: var(--accent-dim); }
-
-/* Main content area */
-.main-content {
-    height: calc(100vh - 48px);
-    overflow: hidden;
-}
-
-/* ─── Chat Layout ─── */
-.chat-layout {
-    display: flex;
-    height: 100%;
-}
-
-/* Sidebar */
-.chat-sidebar {
-    width: var(--sidebar-width);
-    min-width: var(--sidebar-width);
-    background: var(--bg-surface);
-    border-right: 1px solid var(--border);
-    display: flex;
-    flex-direction: column;
-    padding: 12px;
-    gap: 8px;
-    overflow-y: auto;
-    flex-shrink: 0;
-}
-
-.sidebar-header {
-    display: flex;
-    justify-content: space-between;
-    align-items: center;
-}
-
-.sidebar-header h3 { font-size: 0.85em; color: var(--text-secondary); text-transform: uppercase; letter-spacing: 0.05em; }
-
-.sidebar-toggle {
-    display: none;
-    background: none;
-    border: none;
-    color: var(--text-secondary);
-    font-size: 1.2em;
-    cursor: pointer;
-}
-
-.sidebar-open-btn {
-    display: none;
-    position: fixed;
-    top: 56px;
-    left: 8px;
-    z-index: 20;
-    background: var(--bg-surface);
-    border: 1px solid var(--border);
-    color: var(--text-primary);
-    padding: 6px 10px;
-    border-radius: var(--radius);
-    cursor: pointer;
-    font-size: 1.2em;
-}
-
-.sidebar-divider {
-    height: 1px;
-    background: var(--border);
-    margin: 4px 0;
-}
-
-.agent-selector { display: flex; flex-direction: column; gap: 4px; }
-
-.agent-btn {
-    padding: 8px 12px;
-    background: transparent;
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    color: var(--text-primary);
-    cursor: pointer;
-    text-align: left;
-    font-size: 0.9em;
-    transition: border-color 0.15s, background 0.15s;
-}
-.agent-btn:hover { background: var(--bg-surface-hover); }
-.agent-btn.active { border-color: var(--accent); background: var(--accent-dim); }
-
-.btn-new-chat {
-    width: 100%;
-    padding: 8px;
-    background: var(--accent-dim);
-    border: 1px solid var(--accent);
-    border-radius: var(--radius);
-    color: var(--accent);
-    cursor: pointer;
-    font-size: 0.9em;
-    transition: background 0.15s;
-}
-.btn-new-chat:hover { background: var(--accent); color: var(--bg-primary); }
-
-.chat-sidebar h3 {
-    font-size: 0.8em;
-    color: var(--text-secondary);
-    text-transform: uppercase;
-    letter-spacing: 0.05em;
-    margin-top: 8px;
-}
-
-.conv-btn {
-    display: block;
-    width: 100%;
-    padding: 8px 10px;
-    background: transparent;
-    border: 1px solid transparent;
-    border-radius: var(--radius);
-    color: var(--text-primary);
-    cursor: pointer;
-    text-align: left;
-    font-size: 0.85em;
-    white-space: nowrap;
-    overflow: hidden;
-    text-overflow: ellipsis;
-    transition: background 0.15s;
-}
-.conv-btn:hover { background: var(--bg-surface-hover); }
-.conv-btn.active { border-color: var(--accent); background: var(--accent-dim); }
-
-/* Chat main area */
-.chat-main {
-    flex: 1;
-    display: flex;
-    flex-direction: column;
-    min-width: 0;
-}
-
-/* Status bar */
-.status-bar {
-    display: flex;
-    gap: 16px;
-    padding: 8px 20px;
-    font-size: 0.8em;
-    color: var(--text-secondary);
-    border-bottom: 1px solid var(--border);
-    background: var(--bg-surface);
-    flex-shrink: 0;
-}
-.status-item strong { color: var(--text-primary); }
-.text-ok { color: var(--success) !important; }
-.text-err { color: var(--error) !important; }
-
-/* Notification banner */
-.notification-banner {
-    margin: 8px 20px 0;
-    padding: 10px 16px;
-    background: var(--bg-surface);
-    border: 1px solid var(--accent-dim);
-    border-radius: var(--radius);
-    font-size: 0.9em;
-    color: var(--accent);
-}
-
-/* Messages area */
-.chat-messages {
-    flex: 1;
-    overflow-y: auto;
-    padding: 16px 20px;
-    display: flex;
-    flex-direction: column;
-    gap: 12px;
-}
-
-.message {
-    display: flex;
-    gap: 10px;
-    max-width: 85%;
-    animation: fadeIn 0.2s ease-out;
-}
-
-@keyframes fadeIn {
-    from { opacity: 0; transform: translateY(4px); }
-    to { opacity: 1; transform: translateY(0); }
-}
-
-.message.user { align-self: flex-end; flex-direction: row-reverse; }
-.message.assistant { align-self: flex-start; }
-
-.message-avatar {
-    width: 32px;
-    height: 32px;
-    border-radius: 50%;
-    display: flex;
-    align-items: center;
-    justify-content: center;
-    font-size: 0.7em;
-    font-weight: 700;
-    flex-shrink: 0;
-}
-.message.user .message-avatar { background: var(--accent-dim); color: var(--accent); }
-.message.assistant .message-avatar { background: #1c2129; color: var(--text-secondary); }
-
-.message-body {
-    background: var(--bg-surface);
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    padding: 10px 14px;
-    min-width: 0;
-}
-.message.user .message-body { background: var(--accent-dim); border-color: var(--accent); }
-
-.message-content {
-    word-wrap: break-word;
-    overflow-wrap: break-word;
-}
-
-/* Markdown rendering in messages */
-.message-content p { margin: 0.4em 0; }
-.message-content p:first-child { margin-top: 0; }
-.message-content p:last-child { margin-bottom: 0; }
-.message-content pre {
-    background: var(--bg-primary);
-    border: 1px solid var(--border);
-    border-radius: 4px;
-    padding: 10px;
-    overflow-x: auto;
-    font-family: var(--font-mono);
-    font-size: 0.9em;
-    margin: 0.5em 0;
-}
-.message-content code {
-    font-family: var(--font-mono);
-    font-size: 0.9em;
-    background: var(--bg-primary);
-    padding: 2px 5px;
-    border-radius: 3px;
-}
-.message-content pre code { background: none; padding: 0; }
-.message-content ul, .message-content ol { margin: 0.4em 0; padding-left: 1.5em; }
-.message-content a { color: var(--accent); }
-.message-content blockquote {
-    border-left: 3px solid var(--accent);
-    padding-left: 12px;
-    color: var(--text-secondary);
-    margin: 0.5em 0;
-}
-.message-content table { border-collapse: collapse; margin: 0.5em 0; }
-.message-content th, .message-content td {
-    border: 1px solid var(--border);
-    padding: 6px 10px;
-    text-align: left;
-}
-.message-content th { background: var(--bg-surface-hover); }
-
-/* Chat input area */
-.chat-input-area {
-    padding: 12px 20px;
-    border-top: 1px solid var(--border);
-    background: var(--bg-surface);
-    flex-shrink: 0;
-}
-
-.input-row {
-    display: flex;
-    align-items: flex-end;
-    gap: 8px;
-}
-
-#chat-input {
-    flex: 1;
-    background: var(--bg-input);
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    padding: 10px 14px;
-    color: var(--text-primary);
-    font-family: var(--font-sans);
-    font-size: 15px;
-    resize: none;
-    max-height: 200px;
-    line-height: 1.4;
-}
-#chat-input:focus { outline: none; border-color: var(--accent); }
-#chat-input::placeholder { color: var(--text-muted); }
-
-.file-upload-btn {
-    padding: 8px 10px;
-    cursor: pointer;
-    font-size: 1.2em;
-    color: var(--text-secondary);
-    transition: color 0.15s;
-    flex-shrink: 0;
-}
-.file-upload-btn:hover { color: var(--accent); }
-
-.send-btn {
-    padding: 8px 14px;
-    background: var(--accent);
-    border: none;
-    border-radius: var(--radius);
-    color: var(--bg-primary);
-    font-size: 1.1em;
-    cursor: pointer;
-    flex-shrink: 0;
-    transition: opacity 0.15s;
-}
-.send-btn:hover { opacity: 0.85; }
-
-.file-preview {
-    margin-top: 6px;
-    font-size: 0.85em;
-    color: var(--text-secondary);
-}
-.file-preview .file-tag {
-    display: inline-block;
-    background: var(--bg-primary);
-    border: 1px solid var(--border);
-    border-radius: 4px;
-    padding: 2px 8px;
-    margin-right: 6px;
-}
-
-/* ─── Dashboard Layout ─── */
-.dashboard-layout {
-    display: flex;
-    flex-direction: column;
-    gap: 16px;
-    padding: 16px 20px;
-    height: 100%;
-    overflow-y: auto;
-}
-
-.panel {
-    background: var(--bg-surface);
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    padding: 16px;
-}
-
-.panel-title {
-    font-size: 1.1em;
-    font-weight: 600;
-    margin-bottom: 12px;
-    color: var(--accent);
-}
-
-.panel-section {
-    margin-bottom: 16px;
-}
-
-.panel-section h3 {
-    font-size: 0.85em;
-    color: var(--text-secondary);
-    text-transform: uppercase;
-    letter-spacing: 0.05em;
-    margin-bottom: 8px;
-}
-
-/* Loop health grid */
-.loop-grid {
-    display: flex;
-    flex-wrap: wrap;
-    gap: 8px;
-}
-
-.loop-badge {
-    display: flex;
-    flex-direction: column;
-    align-items: center;
-    padding: 8px 12px;
-    border-radius: var(--radius);
-    font-size: 0.8em;
-    min-width: 90px;
-    border: 1px solid var(--border);
-}
-.loop-name { font-weight: 600; }
-.loop-ago { color: var(--text-secondary); font-size: 0.85em; }
-
-.badge-ok { border-color: var(--success); background: rgba(63, 185, 80, 0.1); }
-.badge-ok .loop-name { color: var(--success); }
-.badge-warn { border-color: var(--warning); background: rgba(210, 153, 34, 0.1); }
-.badge-warn .loop-name { color: var(--warning); }
-.badge-err { border-color: var(--error); background: rgba(248, 81, 73, 0.1); }
-.badge-err .loop-name { color: var(--error); }
-.badge-muted { border-color: var(--text-muted); }
-.badge-muted .loop-name { color: var(--text-muted); }
-
-/* Active executions */
-.exec-list { display: flex; flex-direction: column; gap: 6px; }
-.exec-item {
-    display: flex;
-    gap: 12px;
-    padding: 6px 10px;
-    background: var(--bg-primary);
-    border-radius: 4px;
-    font-size: 0.85em;
-}
-.exec-name { flex: 1; font-weight: 500; }
-.exec-tool { color: var(--text-secondary); }
-.exec-dur { color: var(--accent); font-family: var(--font-mono); }
-
-/* Action buttons */
-.action-buttons { display: flex; gap: 8px; flex-wrap: wrap; }
-
-.btn {
-    padding: 8px 16px;
-    background: var(--bg-surface-hover);
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    color: var(--text-primary);
-    cursor: pointer;
-    font-size: 0.9em;
-    transition: border-color 0.15s, background 0.15s;
-}
-.btn:hover { border-color: var(--accent); }
-
-.btn-sm { padding: 6px 12px; font-size: 0.8em; }
-
-/* Notification feed */
-.notif-feed { display: flex; flex-direction: column; gap: 4px; max-height: 300px; overflow-y: auto; }
-.notif-item {
-    padding: 6px 10px;
-    font-size: 0.85em;
-    border-left: 3px solid var(--border);
-    background: var(--bg-primary);
-    border-radius: 0 4px 4px 0;
-}
-.notif-clickup { border-left-color: var(--accent); }
-.notif-info { border-left-color: var(--text-secondary); }
-.notif-error { border-left-color: var(--error); }
-.notif-cat {
-    font-weight: 600;
-    font-size: 0.8em;
-    text-transform: uppercase;
-    color: var(--text-secondary);
-}
-
-/* Task table */
-.task-table { width: 100%; border-collapse: collapse; font-size: 0.85em; }
-.task-table th, .task-table td { padding: 8px 12px; border-bottom: 1px solid var(--border); text-align: left; }
-.task-table th { color: var(--text-secondary); font-weight: 600; text-transform: uppercase; font-size: 0.85em; }
-.task-table a { color: var(--accent); text-decoration: none; }
-.task-table a:hover { text-decoration: underline; }
-
-.status-badge {
-    display: inline-block;
-    padding: 2px 8px;
-    border-radius: 4px;
-    font-size: 0.85em;
-    font-weight: 500;
-}
-.status-to-do { background: rgba(139, 148, 158, 0.2); color: var(--text-secondary); }
-.status-in-progress, .status-automation-underway { background: rgba(45, 212, 191, 0.15); color: var(--accent); }
-.status-error { background: rgba(248, 81, 73, 0.15); color: var(--error); }
-.status-complete, .status-closed { background: rgba(63, 185, 80, 0.15); color: var(--success); }
-.status-internal-review, .status-outline-review { background: rgba(210, 153, 34, 0.15); color: var(--warning); }
-
-/* Pipeline groups */
-.pipeline-group { margin-bottom: 16px; }
-.pipeline-group h4 {
-    font-size: 0.9em;
-    margin-bottom: 8px;
-    padding-bottom: 4px;
-    border-bottom: 1px solid var(--border);
-}
-.pipeline-stats {
-    display: flex;
-    gap: 12px;
-    margin-bottom: 12px;
-    flex-wrap: wrap;
-}
-.pipeline-stat {
-    padding: 8px 14px;
-    background: var(--bg-primary);
-    border: 1px solid var(--border);
-    border-radius: var(--radius);
-    text-align: center;
-}
-.pipeline-stat .stat-count { font-size: 1.5em; font-weight: 700; color: var(--accent); }
-.pipeline-stat .stat-label { font-size: 0.75em; color: var(--text-secondary); }
-
-/* Flash messages */
-.flash-msg {
-    position: fixed;
-    bottom: 20px;
-    right: 20px;
-    background: var(--accent);
-    color: var(--bg-primary);
-    padding: 10px 20px;
-    border-radius: var(--radius);
-    font-weight: 600;
-    font-size: 0.9em;
-    z-index: 100;
-    animation: fadeIn 0.2s ease-out, fadeOut 0.5s 2.5s ease-out forwards;
-}
-@keyframes fadeOut { to { opacity: 0; transform: translateY(10px); } }
-
-/* Utility */
-.text-muted { color: var(--text-muted); }
-
-/* Typing indicator */
-.typing-indicator span {
-    display: inline-block;
-    width: 6px;
-    height: 6px;
-    background: var(--text-secondary);
-    border-radius: 50%;
-    margin: 0 2px;
-    animation: bounce 1.2s infinite;
-}
-.typing-indicator span:nth-child(2) { animation-delay: 0.2s; }
-.typing-indicator span:nth-child(3) { animation-delay: 0.4s; }
-@keyframes bounce {
-    0%, 60%, 100% { transform: translateY(0); }
-    30% { transform: translateY(-6px); }
-}
-
-/* ─── Mobile ─── */
-@media (max-width: 768px) {
-    .chat-sidebar {
-        position: fixed;
-        top: 48px;
-        left: 0;
-        bottom: 0;
-        z-index: 30;
-        transform: translateX(-100%);
-        transition: transform 0.2s ease;
-        width: 280px;
-    }
-    .chat-sidebar.open { transform: translateX(0); }
-    .sidebar-toggle { display: block; }
-    .sidebar-open-btn { display: block; }
-
-    .status-bar { flex-wrap: wrap; gap: 8px; padding: 6px 12px; font-size: 0.75em; }
-
-    .chat-messages { padding: 12px; }
-    .message { max-width: 95%; }
-    .chat-input-area { padding: 8px 12px; }
-
-    #chat-input { font-size: 16px; } /* Prevent iOS zoom */
-
-    .dashboard-layout { padding: 12px; }
-    .loop-grid { gap: 6px; }
-    .loop-badge { min-width: 70px; padding: 6px 8px; font-size: 0.75em; }
-}
-
-/* Overlay for mobile sidebar */
-.sidebar-overlay {
-    display: none;
-    position: fixed;
-    top: 48px;
-    left: 0;
-    right: 0;
-    bottom: 0;
-    background: rgba(0, 0, 0, 0.5);
-    z-index: 25;
-}
-.sidebar-overlay.visible { display: block; }
-
-/* Scrollbar styling */
-::-webkit-scrollbar { width: 6px; }
-::-webkit-scrollbar-track { background: transparent; }
-::-webkit-scrollbar-thumb { background: var(--border); border-radius: 3px; }
-::-webkit-scrollbar-thumb:hover { background: var(--text-muted); }

cheddahbot/static/app.js

@@ -1,284 +0,0 @@
-/* CheddahBot Frontend JS */
-
-// ── Session Management ──
-const SESSION_KEY = 'cheddahbot_session';
-
-function getSession() {
-    try { return JSON.parse(localStorage.getItem(SESSION_KEY) || '{}'); }
-    catch { return {}; }
-}
-
-function saveSession(data) {
-    const s = getSession();
-    Object.assign(s, data);
-    localStorage.setItem(SESSION_KEY, JSON.stringify(s));
-}
-
-function getActiveAgent() {
-    return getSession().agent_name || document.getElementById('input-agent-name')?.value || 'default';
-}
-
-// ── Agent Switching ──
-function switchAgent(name) {
-    // Update UI
-    document.querySelectorAll('.agent-btn').forEach(b => {
-        b.classList.toggle('active', b.dataset.agent === name);
-    });
-    document.getElementById('input-agent-name').value = name;
-    document.getElementById('input-conv-id').value = '';
-    saveSession({ agent_name: name, conv_id: null });
-
-    // Clear chat and load new sidebar
-    document.getElementById('chat-messages').innerHTML = '';
-    refreshSidebar();
-}
-
-function setActiveAgent(name) {
-    document.querySelectorAll('.agent-btn').forEach(b => {
-        b.classList.toggle('active', b.dataset.agent === name);
-    });
-    const agentInput = document.getElementById('input-agent-name');
-    if (agentInput) agentInput.value = name;
-}
-
-// ── Sidebar ──
-function refreshSidebar() {
-    const agent = getActiveAgent();
-    htmx.ajax('GET', '/chat/conversations?agent_name=' + agent, {
-        target: '#sidebar-conversations',
-        swap: 'innerHTML'
-    });
-}
-
-// ── Conversation Loading ──
-function loadConversation(convId) {
-    const agent = getActiveAgent();
-    document.getElementById('input-conv-id').value = convId;
-    saveSession({ conv_id: convId });
-
-    htmx.ajax('GET', '/chat/load/' + convId + '?agent_name=' + agent, {
-        target: '#chat-messages',
-        swap: 'innerHTML'
-    }).then(() => {
-        scrollChat();
-        renderAllMarkdown();
-    });
-}
-
-// ── Chat Input ──
-function handleKeydown(e) {
-    if (e.key === 'Enter' && !e.shiftKey) {
-        e.preventDefault();
-        document.getElementById('chat-form').requestSubmit();
-    }
-}
-
-function autoResize(el) {
-    el.style.height = 'auto';
-    el.style.height = Math.min(el.scrollHeight, 200) + 'px';
-}
-
-function afterSend(event) {
-    const input = document.getElementById('chat-input');
-    input.value = '';
-    input.style.height = 'auto';
-
-    // Clear file input and preview
-    const fileInput = document.querySelector('input[type="file"]');
-    if (fileInput) fileInput.value = '';
-    const preview = document.getElementById('file-preview');
-    if (preview) { preview.style.display = 'none'; preview.innerHTML = ''; }
-
-    scrollChat();
-}
-
-function scrollChat() {
-    const el = document.getElementById('chat-messages');
-    if (el) {
-        requestAnimationFrame(() => {
-            el.scrollTop = el.scrollHeight;
-        });
-    }
-}
-
-// ── File Upload Preview ──
-function showFileNames(input) {
-    const preview = document.getElementById('file-preview');
-    if (!input.files.length) {
-        preview.style.display = 'none';
-        return;
-    }
-    let html = '';
-    for (const f of input.files) {
-        html += '<span class="file-tag">' + f.name + '</span>';
-    }
-    preview.innerHTML = html;
-    preview.style.display = 'block';
-}
-
-// Drag and drop
-document.addEventListener('DOMContentLoaded', () => {
-    const chatMain = document.querySelector('.chat-main');
-    if (!chatMain) return;
-
-    chatMain.addEventListener('dragover', e => {
-        e.preventDefault();
-        chatMain.style.outline = '2px dashed var(--accent)';
-    });
-    chatMain.addEventListener('dragleave', () => {
-        chatMain.style.outline = '';
-    });
-    chatMain.addEventListener('drop', e => {
-        e.preventDefault();
-        chatMain.style.outline = '';
-        const fileInput = document.querySelector('input[type="file"]');
-        if (fileInput && e.dataTransfer.files.length) {
-            fileInput.files = e.dataTransfer.files;
-            showFileNames(fileInput);
-        }
-    });
-});
-
-// ── SSE Streaming ──
-// Handle SSE chunks for chat streaming
-let streamBuffer = '';
-let activeSSE = null;
-
-document.addEventListener('htmx:sseBeforeMessage', function(e) {
-    // This fires for each SSE event received by htmx
-});
-
-// Watch for SSE trigger divs being added to the DOM
-const observer = new MutationObserver(mutations => {
-    for (const m of mutations) {
-        for (const node of m.addedNodes) {
-            if (node.id === 'sse-trigger') {
-                setupStream(node);
-            }
-        }
-    }
-});
-
-document.addEventListener('DOMContentLoaded', () => {
-    const chatMessages = document.getElementById('chat-messages');
-    if (chatMessages) {
-        observer.observe(chatMessages, { childList: true, subtree: true });
-    }
-});
-
-function setupStream(triggerDiv) {
-    const sseUrl = triggerDiv.getAttribute('sse-connect');
-    if (!sseUrl) return;
-
-    // Remove the htmx SSE to manage manually
-    triggerDiv.remove();
-
-    const responseDiv = document.getElementById('assistant-response');
-    if (!responseDiv) return;
-
-    streamBuffer = '';
-
-    // Show typing indicator
-    responseDiv.innerHTML = '<div class="typing-indicator"><span></span><span></span><span></span></div>';
-
-    const source = new EventSource(sseUrl);
-    activeSSE = source;
-
-    source.addEventListener('chunk', function(e) {
-        if (streamBuffer === '') {
-            // Remove typing indicator on first chunk
-            responseDiv.innerHTML = '';
-        }
-        streamBuffer += e.data;
-        // Render markdown
-        try {
-            responseDiv.innerHTML = marked.parse(streamBuffer);
-        } catch {
-            responseDiv.textContent = streamBuffer;
-        }
-        scrollChat();
-    });
-
-    source.addEventListener('done', function(e) {
-        source.close();
-        activeSSE = null;
-        // Final markdown render
-        if (streamBuffer) {
-            try {
-                responseDiv.innerHTML = marked.parse(streamBuffer);
-            } catch {
-                responseDiv.textContent = streamBuffer;
-            }
-        }
-        streamBuffer = '';
-
-        // Update conv_id from done event data
-        const convId = e.data;
-        if (convId) {
-            document.getElementById('input-conv-id').value = convId;
-            saveSession({ conv_id: convId });
-        }
-
-        // Refresh sidebar
-        refreshSidebar();
-        scrollChat();
-    });
-
-    source.onerror = function() {
-        source.close();
-        activeSSE = null;
-        if (!streamBuffer) {
-            responseDiv.innerHTML = '<span class="text-err">Connection lost</span>';
-        }
-    };
-}
-
-// ── Markdown Rendering ──
-function renderAllMarkdown() {
-    document.querySelectorAll('.message-content').forEach(el => {
-        const raw = el.textContent;
-        if (raw && typeof marked !== 'undefined') {
-            try {
-                el.innerHTML = marked.parse(raw);
-            } catch { /* keep raw text */ }
-        }
-    });
-}
-
-// ── Mobile Sidebar ──
-function toggleSidebar() {
-    const sidebar = document.getElementById('chat-sidebar');
-    const overlay = document.getElementById('sidebar-overlay');
-    if (sidebar) {
-        sidebar.classList.toggle('open');
-    }
-    if (overlay) {
-        overlay.classList.toggle('visible');
-    }
-}
-
-// ── Notification Banner (chat page) ──
-function setupChatNotifications() {
-    const banner = document.getElementById('notification-banner');
-    if (!banner) return;
-
-    const source = new EventSource('/sse/notifications');
-    source.addEventListener('notification', function(e) {
-        const notif = JSON.parse(e.data);
-        banner.textContent = notif.message;
-        banner.style.display = 'block';
-        // Auto-hide after 15s
-        setTimeout(() => { banner.style.display = 'none'; }, 15000);
-    });
-}
-
-document.addEventListener('DOMContentLoaded', setupChatNotifications);
-
-// ── HTMX Events ──
-document.addEventListener('scrollChat', scrollChat);
-document.addEventListener('htmx:afterSwap', function(e) {
-    if (e.target.id === 'chat-messages') {
-        renderAllMarkdown();
-        scrollChat();
-    }
-});

cheddahbot/templates/base.html

@@ -1,27 +0,0 @@
-<!DOCTYPE html>
-<html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>{% block title %}CheddahBot{% endblock %}</title>
-    <link rel="stylesheet" href="/static/app.css">
-    <script src="https://unpkg.com/htmx.org@2.0.4"></script>
-    <script src="https://unpkg.com/htmx-ext-sse@2.3.0/sse.js"></script>
-    <script src="https://cdn.jsdelivr.net/npm/marked/marked.min.js"></script>
-    {% block head %}{% endblock %}
-</head>
-<body>
-    <nav class="top-nav">
-        <div class="nav-brand">CheddahBot</div>
-        <div class="nav-links">
-            <a href="/" class="nav-link {% block nav_chat_active %}{% endblock %}">Chat</a>
-            <a href="/dashboard" class="nav-link {% block nav_dash_active %}{% endblock %}">Dashboard</a>
-        </div>
-    </nav>
-    <main class="main-content">
-        {% block content %}{% endblock %}
-    </main>
-    <script src="/static/app.js"></script>
-    {% block scripts %}{% endblock %}
-</body>
-</html>

cheddahbot/templates/chat.html

@@ -1,111 +0,0 @@
-{% extends "base.html" %}
-
-{% block title %}Chat - CheddahBot{% endblock %}
-{% block nav_chat_active %}active{% endblock %}
-
-{% block content %}
-<div class="chat-layout">
-    <!-- Sidebar -->
-    <aside class="chat-sidebar" id="chat-sidebar">
-        <div class="sidebar-header">
-            <h3>Agents</h3>
-            <button class="sidebar-toggle" onclick="toggleSidebar()" aria-label="Close sidebar">&#x2715;</button>
-        </div>
-        <div class="agent-selector" id="agent-selector">
-            {% for agent in agents %}
-            <button
-                class="agent-btn {% if agent.name == default_agent %}active{% endif %}"
-                data-agent="{{ agent.name }}"
-                onclick="switchAgent('{{ agent.name }}')"
-            >{{ agent.display_name }}</button>
-            {% endfor %}
-        </div>
-
-        <div class="sidebar-divider"></div>
-
-        <button
-            class="btn btn-new-chat"
-            hx-post="/chat/new"
-            hx-vals='{"agent_name": "{{ default_agent }}"}'
-            hx-target="#chat-messages"
-            hx-swap="innerHTML"
-            onclick="this.setAttribute('hx-vals', JSON.stringify({agent_name: getActiveAgent()}))"
-        >+ New Chat</button>
-
-        <h3>History</h3>
-        <div id="sidebar-conversations"
-             hx-get="/chat/conversations?agent_name={{ default_agent }}"
-             hx-trigger="load"
-             hx-swap="innerHTML">
-        </div>
-    </aside>
-
-    <!-- Mobile sidebar toggle + overlay -->
-    <button class="sidebar-open-btn" onclick="toggleSidebar()" aria-label="Open sidebar">&#9776;</button>
-    <div id="sidebar-overlay" class="sidebar-overlay" onclick="toggleSidebar()"></div>
-
-    <!-- Chat area -->
-    <div class="chat-main">
-        <!-- Status bar -->
-        <div class="status-bar">
-            <span class="status-item">Model: <strong>{{ chat_model }}</strong></span>
-            <span class="status-item">Exec: <strong class="{% if exec_available %}text-ok{% else %}text-err{% endif %}">{{ "OK" if exec_available else "N/A" }}</strong></span>
-            <span class="status-item">ClickUp: <strong class="{% if clickup_enabled %}text-ok{% else %}text-err{% endif %}">{{ "ON" if clickup_enabled else "OFF" }}</strong></span>
-        </div>
-
-        <!-- Notification banner (populated by SSE) -->
-        <div id="notification-banner" class="notification-banner" style="display:none;"></div>
-
-        <!-- Messages -->
-        <div class="chat-messages" id="chat-messages">
-            <!-- Messages loaded here -->
-        </div>
-
-        <!-- Input area -->
-        <form id="chat-form" class="chat-input-area"
-              hx-post="/chat/send"
-              hx-target="#chat-messages"
-              hx-swap="beforeend"
-              hx-encoding="multipart/form-data"
-              hx-on::after-request="afterSend(event)">
-            <input type="hidden" name="agent_name" id="input-agent-name" value="{{ default_agent }}">
-            <input type="hidden" name="conv_id" id="input-conv-id" value="">
-            <div class="input-row">
-                <label class="file-upload-btn" title="Attach files">
-                    &#x1f4ce;
-                    <input type="file" name="files" multiple style="display:none;" onchange="showFileNames(this)">
-                </label>
-                <textarea name="text" id="chat-input" rows="1" placeholder="Type a message..."
-                          onkeydown="handleKeydown(event)" oninput="autoResize(this)"></textarea>
-                <button type="submit" class="send-btn" title="Send">&#x27A4;</button>
-            </div>
-            <div id="file-preview" class="file-preview" style="display:none;"></div>
-        </form>
-    </div>
-</div>
-{% endblock %}
-
-{% block scripts %}
-<script>
-    // Initialize session state
-    const SESSION_KEY = 'cheddahbot_session';
-    let session = JSON.parse(localStorage.getItem(SESSION_KEY) || '{}');
-    if (!session.agent_name) session.agent_name = '{{ default_agent }}';
-
-    // Restore session on load
-    document.addEventListener('DOMContentLoaded', function() {
-        if (session.agent_name) {
-            setActiveAgent(session.agent_name);
-        }
-        if (session.conv_id) {
-            loadConversation(session.conv_id);
-        }
-        // Load conversations for sidebar
-        refreshSidebar();
-    });
-
-    function saveSession() {
-        localStorage.setItem(SESSION_KEY, JSON.stringify(session));
-    }
-</script>
-{% endblock %}

cheddahbot/templates/dashboard.html

@@ -1,174 +0,0 @@
-{% extends "base.html" %}
-
-{% block title %}Dashboard - CheddahBot{% endblock %}
-{% block nav_dash_active %}active{% endblock %}
-
-{% block content %}
-<div class="dashboard-layout">
-
-    <!-- Ops Panel -->
-    <section class="panel" id="ops-panel">
-        <h2 class="panel-title">Operations</h2>
-
-        <!-- Active Executions -->
-        <div class="panel-section">
-            <h3>Active Executions</h3>
-            <div id="active-executions" class="exec-list">
-                <span class="text-muted">Loading...</span>
-            </div>
-        </div>
-
-        <!-- Loop Health -->
-        <div class="panel-section">
-            <h3>Loop Health</h3>
-            <div id="loop-health" class="loop-grid">
-                <span class="text-muted">Loading...</span>
-            </div>
-        </div>
-
-        <!-- Actions -->
-        <div class="panel-section">
-            <h3>Actions</h3>
-            <div class="action-buttons">
-                <button class="btn btn-sm"
-                    hx-post="/api/system/loops/force"
-                    hx-swap="none"
-                    hx-on::after-request="showFlash('Force pulse sent')">
-                    Force Pulse
-                </button>
-                <button class="btn btn-sm"
-                    hx-post="/api/system/briefing/force"
-                    hx-swap="none"
-                    hx-on::after-request="showFlash('Briefing triggered')">
-                    Force Briefing
-                </button>
-                <button class="btn btn-sm"
-                    hx-post="/api/cache/clear"
-                    hx-swap="none"
-                    hx-on::after-request="showFlash('Cache cleared')">
-                    Clear Cache
-                </button>
-            </div>
-        </div>
-
-        <!-- Notification Feed -->
-        <div class="panel-section">
-            <h3>Notifications</h3>
-            <div id="notification-feed" class="notif-feed">
-                <span class="text-muted">Waiting for notifications...</span>
-            </div>
-        </div>
-    </section>
-
-    <!-- Pipeline Panel -->
-    <section class="panel" id="pipeline-panel">
-        <h2 class="panel-title">Pipeline</h2>
-        <div id="pipeline-content"
-             hx-get="/dashboard/pipeline"
-             hx-trigger="load, every 120s"
-             hx-swap="innerHTML">
-            <span class="text-muted">Loading pipeline data...</span>
-        </div>
-    </section>
-
-</div>
-{% endblock %}
-
-{% block scripts %}
-<script>
-    // Connect to SSE for live loop updates
-    const loopSource = new EventSource('/sse/loops');
-    loopSource.addEventListener('loops', function(e) {
-        const data = JSON.parse(e.data);
-        renderLoopHealth(data.loops);
-        renderActiveExecutions(data.executions);
-    });
-
-    // Connect to SSE for notifications
-    const notifSource = new EventSource('/sse/notifications');
-    notifSource.addEventListener('notification', function(e) {
-        const notif = JSON.parse(e.data);
-        addNotification(notif.message, notif.category);
-    });
-
-    function renderLoopHealth(loops) {
-        const container = document.getElementById('loop-health');
-        if (!loops || Object.keys(loops).length === 0) {
-            container.innerHTML = '<span class="text-muted">No loop data</span>';
-            return;
-        }
-        let html = '';
-        const now = new Date();
-        for (const [name, ts] of Object.entries(loops)) {
-            let statusClass = 'badge-muted';
-            let agoText = 'never';
-            if (ts) {
-                const dt = new Date(ts);
-                const secs = Math.floor((now - dt) / 1000);
-                if (secs < 120) {
-                    statusClass = 'badge-ok';
-                    agoText = secs + 's ago';
-                } else if (secs < 600) {
-                    statusClass = 'badge-warn';
-                    agoText = Math.floor(secs / 60) + 'm ago';
-                } else {
-                    statusClass = 'badge-err';
-                    agoText = Math.floor(secs / 60) + 'm ago';
-                }
-            }
-            html += '<div class="loop-badge ' + statusClass + '">' +
-                    '<span class="loop-name">' + name + '</span>' +
-                    '<span class="loop-ago">' + agoText + '</span>' +
-                    '</div>';
-        }
-        container.innerHTML = html;
-    }
-
-    function renderActiveExecutions(execs) {
-        const container = document.getElementById('active-executions');
-        if (!execs || Object.keys(execs).length === 0) {
-            container.innerHTML = '<span class="text-muted">No active executions</span>';
-            return;
-        }
-        let html = '';
-        const now = new Date();
-        for (const [id, info] of Object.entries(execs)) {
-            const started = new Date(info.started_at);
-            const durSecs = Math.floor((now - started) / 1000);
-            let dur = durSecs + 's';
-            if (durSecs >= 60) dur = Math.floor(durSecs / 60) + 'm ' + (durSecs % 60) + 's';
-            html += '<div class="exec-item">' +
-                    '<span class="exec-name">' + info.name + '</span>' +
-                    '<span class="exec-tool">' + info.tool + '</span>' +
-                    '<span class="exec-dur">' + dur + '</span>' +
-                    '</div>';
-        }
-        container.innerHTML = html;
-    }
-
-    let notifCount = 0;
-    function addNotification(message, category) {
-        const container = document.getElementById('notification-feed');
-        if (notifCount === 0) container.innerHTML = '';
-        notifCount++;
-
-        const div = document.createElement('div');
-        div.className = 'notif-item notif-' + (category || 'info');
-        div.innerHTML = '<span class="notif-cat">' + (category || 'info') + '</span> ' + message;
-        container.insertBefore(div, container.firstChild);
-
-        // Keep max 30
-        while (container.children.length > 30) {
-            container.removeChild(container.lastChild);
-        }
-    }
-
-    function showFlash(msg) {
-        const el = document.createElement('div');
-        el.className = 'flash-msg';
-        el.textContent = msg;
-        document.body.appendChild(el);
-        setTimeout(() => el.remove(), 3000);
-    }
-</script>
-{% endblock %}

cheddahbot/templates/partials/chat_message.html

@@ -1,6 +0,0 @@
-<div class="message {{ role }}">
-    <div class="message-avatar">{% if role == 'user' %}You{% else %}CB{% endif %}</div>
-    <div class="message-body">
-        <div class="message-content">{{ content }}</div>
-    </div>
-</div>

cheddahbot/templates/partials/chat_sidebar.html

@@ -1,11 +0,0 @@
-{% if conversations %}
-{% for conv in conversations %}
-<button class="conv-btn"
-        onclick="loadConversation('{{ conv.id }}')"
-        title="{{ conv.title or 'New Chat' }}">
-    {{ conv.title or 'New Chat' }}
-</button>
-{% endfor %}
-{% else %}
-<p class="text-muted">No conversations yet</p>
-{% endif %}

cheddahbot/templates/partials/loop_status.html

@@ -1,6 +0,0 @@
-{% for name, info in loops.items() %}
-<div class="loop-badge {{ info.class }}">
-    <span class="loop-name">{{ name }}</span>
-    <span class="loop-ago">{{ info.ago }}</span>
-</div>
-{% endfor %}

cheddahbot/templates/partials/notification_feed.html

@@ -1,6 +0,0 @@
-{% for notif in notifications %}
-<div class="notif-item notif-{{ notif.category or 'info' }}">
-    <span class="notif-cat">{{ notif.category }}</span>
-    {{ notif.message }}
-</div>
-{% endfor %}

cheddahbot/templates/partials/task_table.html

@@ -1,27 +0,0 @@
-{% if tasks %}
-<table class="task-table">
-    <thead>
-        <tr>
-            <th>Task</th>
-            <th>Customer</th>
-            <th>Status</th>
-            <th>Due</th>
-        </tr>
-    </thead>
-    <tbody>
-        {% for task in tasks %}
-        <tr>
-            <td>
-                {% if task.url %}<a href="{{ task.url }}" target="_blank" rel="noopener">{{ task.name }}</a>
-                {% else %}{{ task.name }}{% endif %}
-            </td>
-            <td>{{ task.custom_fields.get('Client', 'N/A') if task.custom_fields else 'N/A' }}</td>
-            <td><span class="status-badge status-{{ task.status|replace(' ', '-') }}">{{ task.status }}</span></td>
-            <td>{{ task.due_display or '-' }}</td>
-        </tr>
-        {% endfor %}
-    </tbody>
-</table>
-{% else %}
-<p class="text-muted">No tasks</p>
-{% endif %}

cheddahbot/tools/__init__.py

@@ -105,7 +105,6 @@ class ToolRegistry:
         self.db = db
         self.agent = agent
         self.agent_registry = None  # set after multi-agent setup
-        self.scheduler = None  # set after scheduler creation
         self._discover_tools()
 
     def _discover_tools(self):
@@ -159,7 +158,6 @@ class ToolRegistry:
                     "agent": self.agent,
                     "memory": self.agent._memory,
                     "agent_registry": self.agent_registry,
-                    "scheduler": self.scheduler,
                 }
                 # Pass scheduler-injected metadata through ctx (not LLM-visible)
                 if "clickup_task_id" in args:

cheddahbot/tools/autocora.py

@@ -347,7 +347,7 @@ def submit_autocora_jobs(target_date: str = "", ctx: dict | None = None) -> str:
             client.update_task_status(tid, "automation underway")
 
         submitted.append(group["keyword"])
-        log.info("Submitted AutoCora job: %s -> %s", group["keyword"], job_id)
+        log.info("Submitted AutoCora job: %s → %s", group["keyword"], job_id)
 
     # Build response
     lines = [f"AutoCora submission ({label}):"]

cheddahbot/tools/clickup_tool.py

@@ -3,7 +3,6 @@
 from __future__ import annotations
 
 import logging
-from datetime import UTC, datetime
 
 from . import tool
 
@@ -285,79 +284,3 @@ def clickup_reset_task(task_id: str, ctx: dict | None = None) -> str:
         f"Task '{task_id}' reset to '{reset_status}'. "
         f"It will be picked up on the next scheduler poll."
     )
-
-
-def _format_duration(delta) -> str:
-    """Format a timedelta as a human-readable duration string."""
-    total_seconds = int(delta.total_seconds())
-    hours, remainder = divmod(total_seconds, 3600)
-    minutes, seconds = divmod(remainder, 60)
-    if hours:
-        return f"{hours}h {minutes}m {seconds}s"
-    if minutes:
-        return f"{minutes}m {seconds}s"
-    return f"{seconds}s"
-
-
-def _format_ago(iso_str: str | None) -> str:
-    """Format an ISO timestamp as 'Xm ago' relative to now."""
-    if not iso_str:
-        return "never"
-    try:
-        ts = datetime.fromisoformat(iso_str)
-        delta = datetime.now(UTC) - ts
-        total_seconds = int(delta.total_seconds())
-        if total_seconds < 60:
-            return f"{total_seconds}s ago"
-        minutes = total_seconds // 60
-        if minutes < 60:
-            return f"{minutes}m ago"
-        hours = minutes // 60
-        return f"{hours}h {minutes % 60}m ago"
-    except (ValueError, TypeError):
-        return "unknown"
-
-
-@tool(
-    "get_active_tasks",
-    "Show what CheddahBot is actively executing right now. "
-    "Reports running tasks, loop health, and whether it's safe to restart.",
-    category="clickup",
-)
-def get_active_tasks(ctx: dict | None = None) -> str:
-    """Show actively running scheduler tasks and loop health."""
-    scheduler = ctx.get("scheduler") if ctx else None
-    if not scheduler:
-        return "Scheduler not available — cannot check active executions."
-
-    now = datetime.now(UTC)
-    lines = []
-
-    # Active executions
-    active = scheduler.get_active_executions()
-    if active:
-        lines.append(f"**Active Executions ({len(active)}):**")
-        for task_id, info in active.items():
-            duration = _format_duration(now - info["started_at"])
-            lines.append(
-                f"- **{info['name']}** — `{info['tool']}` — "
-                f"running {duration} ({info['thread']} thread)"
-            )
-    else:
-        lines.append("**No tasks actively executing.**")
-
-    # Loop health
-    timestamps = scheduler.get_loop_timestamps()
-    lines.append("")
-    lines.append("**Loop Health:**")
-    for loop_name, ts in timestamps.items():
-        lines.append(f"- {loop_name}: last ran {_format_ago(ts)}")
-
-    # Safe to restart?
-    lines.append("")
-    if active:
-        lines.append(f"**Safe to restart: No** ({len(active)} task(s) actively running)")
-    else:
-        lines.append("**Safe to restart: Yes**")
-
-    return "\n".join(lines)

cheddahbot/tools/content_creation.py

@@ -77,7 +77,7 @@ def _sync_clickup_outline_ready(ctx: dict | None, task_id: str, outline_path: st
 
         client.add_comment(
             task_id,
-            f"[OUTLINE]CheddahBot generated a content outline.\n\n"
+            f"📝 CheddahBot generated a content outline.\n\n"
             f"Outline saved to: `{outline_path}`\n\n"
             f"Please review and edit the outline, then move this task to "
             f"**outline approved** to trigger the full content write.",
@@ -100,7 +100,7 @@ def _sync_clickup_complete(ctx: dict | None, task_id: str, content_path: str) ->
         config = ctx["config"]
         client.add_comment(
             task_id,
-            f"[DONE]CheddahBot completed the content.\n\n"
+            f"✅ CheddahBot completed the content.\n\n"
             f"Final content saved to: `{content_path}`\n\n"
             f"Ready for internal review.",
         )
@@ -122,7 +122,7 @@ def _sync_clickup_fail(ctx: dict | None, task_id: str, error: str) -> None:
         config = ctx["config"]
         client.add_comment(
             task_id,
-            f"[FAILED]CheddahBot failed during content creation.\n\nError: {error[:2000]}",
+            f"❌ CheddahBot failed during content creation.\n\nError: {error[:2000]}",
         )
         client.update_task_status(task_id, config.clickup.error_status)
     except Exception as e:
@@ -489,7 +489,7 @@ def _build_optimization_prompt(
         f"Generate `{work_dir}/optimization_instructions.md` — a surgical playbook "
         f"for the human editor with these sections:\n\n"
         f"1. **Executive Summary** — one-paragraph overview of optimization opportunity\n"
-        f"2. **Heading Changes** — specific H1/H2/H3 modifications with before/after\n"
+        f"2. **Heading Changes** — specific H1/H2/H3 modifications with before→after\n"
         f"3. **Sections to Expand** — which sections need more content and what to add\n"
         f"4. **Entity Integration Points** — exact locations to weave in missing entities\n"
         f"5. **Meta Tag Updates** — title tag and meta description recommendations\n"
@@ -645,7 +645,7 @@ def _finalize_optimization(
             for name, fpath in found_files.items():
                 dest = net_dir / name
                 dest.write_bytes(fpath.read_bytes())
-                log.info("Copied %s -> %s", fpath, dest)
+                log.info("Copied %s → %s", fpath, dest)
         except OSError as e:
             log.warning("Could not copy deliverables to network path %s: %s", net_dir, e)
 
@@ -699,7 +699,7 @@ def _sync_clickup_optimization_complete(
 
         # Build comment with validation summary
         comment_parts = [
-            f"[DONE]Optimization pipeline complete for '{keyword}'.\n",
+            f"✅ Optimization pipeline complete for '{keyword}'.\n",
             f"**URL:** {url}\n",
             "**Deliverables attached:**",
         ]
@@ -1018,7 +1018,7 @@ def _run_phase2(
                     client.update_task_status(task_id, reset_status)
                     client.add_comment(
                         task_id,
-                        f"[WARNING]Outline file not found for keyword '{keyword}'. "
+                        f"⚠️ Outline file not found for keyword '{keyword}'. "
                         f"Searched: {outline_path or '(no path saved)'}. "
                         f"Please re-run Phase 1 (create_content) to generate a new outline.",
                     )

cheddahbot/tools/linkbuilding.py

@@ -10,7 +10,6 @@ import logging
 import os
 import re
 import subprocess
-from collections.abc import Callable
 from pathlib import Path
 
 from . import tool
@@ -30,13 +29,6 @@ def _get_blm_dir(ctx: dict | None) -> str:
     return os.getenv("BLM_DIR", "E:/dev/Big-Link-Man")
 
 
-def _get_blm_timeout(ctx: dict | None) -> int:
-    """Get BLM subprocess timeout from config or default (1800s / 30 min)."""
-    if ctx and "config" in ctx:
-        return ctx["config"].timeouts.blm
-    return 1800
-
-
 def _run_blm_command(
     args: list[str], blm_dir: str, timeout: int = 1800
 ) -> subprocess.CompletedProcess:
@@ -272,20 +264,26 @@ def _normalize_for_match(text: str) -> str:
     return text
 
 
-def _fuzzy_keyword_match(a: str, b: str, llm_check: Callable[[str, str], bool] | None = None) -> bool:
+def _fuzzy_keyword_match(a: str, b: str) -> bool:
-    """Check if two normalized strings match, allowing singular/plural differences.
+    """Check if two normalized strings are a fuzzy match.
 
-    Fast path: exact match after normalization.
+    Matches if: exact, substring in either direction, or >80% word overlap.
-    Slow path: ask an LLM if the two keywords are the same aside from plural form.
-    Falls back to False if no llm_check is provided and strings differ.
     """
     if not a or not b:
         return False
     if a == b:
         return True
-    if llm_check is None:
+    if a in b or b in a:
+        return True
+
+    # Word overlap check
+    words_a = set(a.split())
+    words_b = set(b.split())
+    if not words_a or not words_b:
         return False
-    return llm_check(a, b)
+    overlap = len(words_a & words_b)
+    min_len = min(len(words_a), len(words_b))
+    return overlap / min_len >= 0.8 if min_len > 0 else False
 
 
 def _complete_clickup_task(ctx: dict | None, task_id: str, message: str, status: str = "") -> None:
@@ -322,7 +320,7 @@ def _fail_clickup_task(ctx: dict | None, task_id: str, error_msg: str) -> None:
         try:
             cu_client.add_comment(
                 task_id,
-                f"[FAILED]Link building pipeline failed.\n\nError: {error_msg[:2000]}",
+                f"❌ Link building pipeline failed.\n\nError: {error_msg[:2000]}",
             )
             cu_client.update_task_status(task_id, error_status)
         except Exception as e:
@@ -436,7 +434,7 @@ def run_cora_backlinks(
     # ── Step 1: ingest-cora ──
     _set_status(ctx, f"Step 1/2: Ingesting CORA report for {project_name}...")
     if clickup_task_id:
-        _sync_clickup(ctx, clickup_task_id, "ingest", "[STARTED]Starting Cora Backlinks pipeline...")
+        _sync_clickup(ctx, clickup_task_id, "ingest", "🔄 Starting Cora Backlinks pipeline...")
 
     # Convert branded_plus_ratio from string if needed
     try:
@@ -453,11 +451,10 @@ def run_cora_backlinks(
         cli_flags=cli_flags,
     )
 
-    blm_timeout = _get_blm_timeout(ctx)
     try:
-        ingest_result = _run_blm_command(ingest_args, blm_dir, timeout=blm_timeout)
+        ingest_result = _run_blm_command(ingest_args, blm_dir)
     except subprocess.TimeoutExpired:
-        error = f"ingest-cora timed out after {blm_timeout // 60} minutes"
+        error = "ingest-cora timed out after 30 minutes"
         _set_status(ctx, "")
         if clickup_task_id:
             _fail_clickup_task(ctx, clickup_task_id, error)
@@ -490,7 +487,7 @@ def run_cora_backlinks(
             ctx,
             clickup_task_id,
             "ingest_done",
-            f"[DONE]CORA report ingested. Project ID: {project_id}. Job file: {job_file}",
+            f"✅ CORA report ingested. Project ID: {project_id}. Job file: {job_file}",
         )
 
     # ── Step 2: generate-batch ──
@@ -502,9 +499,9 @@ def run_cora_backlinks(
     gen_args = ["generate-batch", "-j", str(job_path), "--continue-on-error"]
 
     try:
-        gen_result = _run_blm_command(gen_args, blm_dir, timeout=blm_timeout)
+        gen_result = _run_blm_command(gen_args, blm_dir)
     except subprocess.TimeoutExpired:
-        error = f"generate-batch timed out after {blm_timeout // 60} minutes"
+        error = "generate-batch timed out after 30 minutes"
         _set_status(ctx, "")
         if clickup_task_id:
             _fail_clickup_task(ctx, clickup_task_id, error)
@@ -534,7 +531,7 @@ def run_cora_backlinks(
 
     if clickup_task_id:
         summary = (
-            f"[DONE]Cora Backlinks pipeline completed for {project_name}.\n\n"
+            f"✅ Cora Backlinks pipeline completed for {project_name}.\n\n"
             f"Project ID: {project_id}\n"
             f"Keyword: {ingest_parsed['main_keyword']}\n"
             f"Job file: {gen_parsed['job_moved_to'] or job_file}"
@@ -592,11 +589,10 @@ def blm_ingest_cora(
         cli_flags=cli_flags,
     )
 
-    blm_timeout = _get_blm_timeout(ctx)
     try:
-        result = _run_blm_command(ingest_args, blm_dir, timeout=blm_timeout)
+        result = _run_blm_command(ingest_args, blm_dir)
     except subprocess.TimeoutExpired:
-        return f"Error: ingest-cora timed out after {blm_timeout // 60} minutes."
+        return "Error: ingest-cora timed out after 30 minutes."
 
     parsed = _parse_ingest_output(result.stdout)
 
@@ -647,11 +643,10 @@ def blm_generate_batch(
     if debug:
         args.append("--debug")
 
-    blm_timeout = _get_blm_timeout(ctx)
     try:
-        result = _run_blm_command(args, blm_dir, timeout=blm_timeout)
+        result = _run_blm_command(args, blm_dir)
     except subprocess.TimeoutExpired:
-        return f"Error: generate-batch timed out after {blm_timeout // 60} minutes."
+        return "Error: generate-batch timed out after 30 minutes."
 
     parsed = _parse_generate_output(result.stdout)
 

cheddahbot/tools/press_release.py

@@ -574,7 +574,7 @@ def write_press_releases(
                 cu_client.update_task_status(clickup_task_id, config.clickup.automation_status)
                 cu_client.add_comment(
                     clickup_task_id,
-                    f"[STARTED]CheddahBot starting press release creation.\n\n"
+                    f"🔄 CheddahBot starting press release creation.\n\n"
                     f"Topic: {topic}\nCompany: {company_name}",
                 )
                 log.info("ClickUp task %s set to automation-underway", clickup_task_id)
@@ -752,7 +752,7 @@ def write_press_releases(
             if failed_uploads:
                 paths_list = "\n".join(f"  - {p}" for p in failed_uploads)
                 upload_warning = (
-                    f"\n[WARNING]Warning: {len(failed_uploads)} attachment(s) failed to upload. "
+                    f"\n⚠️ Warning: {len(failed_uploads)} attachment(s) failed to upload. "
                     f"Files saved locally at:\n{paths_list}"
                 )
             cu_client.add_comment(
@@ -855,7 +855,7 @@ def write_press_releases(
             attach_note = f"\n📎 {uploaded_count} file(s) attached." if uploaded_count else ""
             result_text = "\n".join(output_parts)[:3000]
             comment = (
-                f"[DONE]CheddahBot completed this task.\n\n"
+                f"✅ CheddahBot completed this task.\n\n"
                 f"Skill: write_press_releases\n"
                 f"Result:\n{result_text}{attach_note}"
             )
@@ -1270,7 +1270,7 @@ def submit_press_release(
     if link_list:
         output_parts.append("\n**Links:**")
         for link in link_list:
-            output_parts.append(f'  - "{link["anchor"]}" -> {link["url"]}')
+            output_parts.append(f'  - "{link["anchor"]}" → {link["url"]}')
 
     if link_warnings:
         output_parts.append("\n**Link warnings:**")

cheddahbot/web/__init__.py

@@ -1,57 +0,0 @@
-"""HTMX + FastAPI web frontend for CheddahBot."""
-
-from __future__ import annotations
-
-import logging
-from pathlib import Path
-from typing import TYPE_CHECKING
-
-from fastapi import FastAPI
-from fastapi.templating import Jinja2Templates
-from starlette.staticfiles import StaticFiles
-
-if TYPE_CHECKING:
-    from ..agent_registry import AgentRegistry
-    from ..config import Config
-    from ..db import Database
-    from ..llm import LLMAdapter
-    from ..notifications import NotificationBus
-    from ..scheduler import Scheduler
-
-log = logging.getLogger(__name__)
-
-_TEMPLATE_DIR = Path(__file__).resolve().parent.parent / "templates"
-_STATIC_DIR = Path(__file__).resolve().parent.parent / "static"
-
-templates = Jinja2Templates(directory=str(_TEMPLATE_DIR))
-
-
-def mount_web_app(
-    app: FastAPI,
-    registry: AgentRegistry,
-    config: Config,
-    llm: LLMAdapter,
-    notification_bus: NotificationBus | None = None,
-    scheduler: Scheduler | None = None,
-    db: Database | None = None,
-):
-    """Mount all web routes and static files onto the FastAPI app."""
-    # Wire dependencies into route modules
-    from . import routes_chat, routes_pages, routes_sse
-    from .routes_chat import router as chat_router
-    from .routes_pages import router as pages_router
-    from .routes_sse import router as sse_router
-
-    routes_pages.setup(registry, config, llm, templates, db=db, scheduler=scheduler)
-    routes_chat.setup(registry, config, llm, db, templates)
-    routes_sse.setup(notification_bus, scheduler, db)
-
-    app.include_router(chat_router)
-    app.include_router(sse_router)
-    # Pages router last (it has catch-all GET /)
-    app.include_router(pages_router)
-
-    # Static files
-    app.mount("/static", StaticFiles(directory=str(_STATIC_DIR)), name="static")
-
-    log.info("Web UI mounted (templates: %s, static: %s)", _TEMPLATE_DIR, _STATIC_DIR)

cheddahbot/web/routes_chat.py

@@ -1,270 +0,0 @@
-"""Chat routes: send messages, stream responses, manage conversations."""
-
-from __future__ import annotations
-
-import asyncio
-import logging
-import tempfile
-import time
-from pathlib import Path
-from typing import TYPE_CHECKING
-
-from fastapi import APIRouter, Form, Request, UploadFile
-from fastapi.responses import HTMLResponse
-from fastapi.templating import Jinja2Templates
-from sse_starlette.sse import EventSourceResponse
-
-if TYPE_CHECKING:
-    from ..agent_registry import AgentRegistry
-    from ..config import Config
-    from ..db import Database
-    from ..llm import LLMAdapter
-
-log = logging.getLogger(__name__)
-
-router = APIRouter(prefix="/chat")
-
-_registry: AgentRegistry | None = None
-_config: Config | None = None
-_llm: LLMAdapter | None = None
-_db: Database | None = None
-_templates: Jinja2Templates | None = None
-
-# Pending responses: conv_id -> {text, files, timestamp}
-_pending: dict[str, dict] = {}
-
-
-def setup(registry, config, llm, db, templates):
-    global _registry, _config, _llm, _db, _templates
-    _registry = registry
-    _config = config
-    _llm = llm
-    _db = db
-    _templates = templates
-
-
-def _get_agent(name: str):
-    if _registry:
-        return _registry.get(name) or _registry.default
-    return None
-
-
-def _cleanup_pending():
-    """Remove pending entries older than 60s."""
-    now = time.time()
-    expired = [k for k, v in _pending.items() if now - v["timestamp"] > 60]
-    for k in expired:
-        del _pending[k]
-
-
-@router.post("/send")
-async def send_message(
-    request: Request,
-    text: str = Form(""),
-    agent_name: str = Form("default"),
-    conv_id: str = Form(""),
-    files: list[UploadFile] | None = None,
-):
-    """Accept user message, return user bubble HTML + trigger SSE stream."""
-    _cleanup_pending()
-
-    agent = _get_agent(agent_name)
-    if not agent:
-        return HTMLResponse("<div class='error'>Agent not found</div>", status_code=400)
-
-    # Handle file uploads
-    saved_files = []
-    for f in (files or []):
-        if f.filename and f.size and f.size > 0:
-            tmp = Path(tempfile.mkdtemp()) / f.filename
-            content = await f.read()
-            tmp.write_bytes(content)
-            saved_files.append(str(tmp))
-
-    if not text.strip() and not saved_files:
-        return HTMLResponse("")
-
-    # Ensure conversation exists
-    if not conv_id:
-        agent.new_conversation()
-        conv_id = agent.ensure_conversation()
-    else:
-        agent.conv_id = conv_id
-
-    # Build display text
-    display_text = text
-    if saved_files:
-        file_names = [Path(f).name for f in saved_files]
-        display_text += f"\n[Attached: {', '.join(file_names)}]"
-
-    # Stash for SSE stream
-    _pending[conv_id] = {
-        "text": text,
-        "files": saved_files,
-        "timestamp": time.time(),
-        "agent_name": agent_name,
-    }
-
-    # Render user bubble + SSE trigger div
-    user_html = _templates.get_template("partials/chat_message.html").render(
-        role="user", content=display_text
-    )
-    # The SSE trigger div connects to the stream endpoint
-    sse_div = (
-        f'<div id="sse-trigger" '
-        f'hx-ext="sse" '
-        f'sse-connect="/chat/stream/{conv_id}" '
-        f'sse-swap="chunk" '
-        f'hx-target="#assistant-response" '
-        f'hx-swap="beforeend">'
-        f'</div>'
-        f'<div id="assistant-bubble" class="message assistant">'
-        f'<div class="message-avatar">CB</div>'
-        f'<div class="message-body">'
-        f'<div id="assistant-response" class="message-content"></div>'
-        f'</div></div>'
-    )
-
-    headers = {
-        "HX-Trigger-After-Swap": "scrollChat",
-        "HX-Push-Url": f"/?conv={conv_id}",
-    }
-
-    return HTMLResponse(user_html + sse_div, headers=headers)
-
-
-@router.get("/stream/{conv_id}")
-async def stream_response(conv_id: str):
-    """SSE endpoint: stream assistant response chunks."""
-    pending = _pending.pop(conv_id, None)
-    if not pending:
-        async def empty():
-            yield {"event": "done", "data": ""}
-        return EventSourceResponse(empty())
-
-    agent = _get_agent(pending["agent_name"])
-    if not agent:
-        async def error():
-            yield {"event": "chunk", "data": "Agent not found"}
-            yield {"event": "done", "data": ""}
-        return EventSourceResponse(error())
-
-    agent.conv_id = conv_id
-
-    async def generate():
-        loop = asyncio.get_event_loop()
-        queue: asyncio.Queue = asyncio.Queue()
-
-        def run_agent():
-            try:
-                for chunk in agent.respond(pending["text"], files=pending.get("files")):
-                    loop.call_soon_threadsafe(queue.put_nowait, ("chunk", chunk))
-            except Exception as e:
-                log.error("Stream error: %s", e, exc_info=True)
-                loop.call_soon_threadsafe(
-                    queue.put_nowait, ("chunk", f"\n\nError: {e}")
-                )
-            finally:
-                loop.call_soon_threadsafe(queue.put_nowait, ("done", ""))
-
-        # Run agent.respond() in a thread
-        import threading
-        t = threading.Thread(target=run_agent, daemon=True)
-        t.start()
-
-        while True:
-            event, data = await queue.get()
-            if event == "done":
-                yield {"event": "done", "data": conv_id}
-                break
-            yield {"event": "chunk", "data": data}
-
-    return EventSourceResponse(generate())
-
-
-@router.get("/conversations")
-async def list_conversations(agent_name: str = "default"):
-    """Return sidebar conversation list as HTML partial."""
-    agent = _get_agent(agent_name)
-    if not agent:
-        return HTMLResponse("")
-
-    convs = agent.db.list_conversations(limit=50, agent_name=agent_name)
-    html = _templates.get_template("partials/chat_sidebar.html").render(
-        conversations=convs
-    )
-    return HTMLResponse(html)
-
-
-@router.post("/new")
-async def new_conversation(agent_name: str = Form("default")):
-    """Create a new conversation, return empty chat + updated sidebar."""
-    agent = _get_agent(agent_name)
-    if not agent:
-        return HTMLResponse("")
-
-    agent.new_conversation()
-    conv_id = agent.ensure_conversation()
-
-    convs = agent.db.list_conversations(limit=50, agent_name=agent_name)
-    sidebar_html = _templates.get_template("partials/chat_sidebar.html").render(
-        conversations=convs
-    )
-
-    # Return empty chat area + sidebar update via OOB swap
-    html = (
-        f'<div id="chat-messages"></div>'
-        f'<div id="sidebar-conversations" hx-swap-oob="innerHTML">'
-        f'{sidebar_html}</div>'
-    )
-
-    headers = {"HX-Push-Url": f"/?conv={conv_id}"}
-    return HTMLResponse(html, headers=headers)
-
-
-@router.get("/load/{conv_id}")
-async def load_conversation(conv_id: str, agent_name: str = "default"):
-    """Load conversation history as HTML."""
-    agent = _get_agent(agent_name)
-    if not agent:
-        return HTMLResponse("")
-
-    messages = agent.load_conversation(conv_id)
-    parts = []
-    for msg in messages:
-        role = msg.get("role", "")
-        content = msg.get("content", "")
-        if role in ("user", "assistant") and content:
-            parts.append(
-                _templates.get_template("partials/chat_message.html").render(
-                    role=role, content=content
-                )
-            )
-
-    headers = {"HX-Push-Url": f"/?conv={conv_id}"}
-    return HTMLResponse("\n".join(parts), headers=headers)
-
-
-@router.post("/agent/{name}")
-async def switch_agent(name: str):
-    """Switch active agent. Returns updated sidebar via OOB."""
-    agent = _get_agent(name)
-    if not agent:
-        return HTMLResponse("<div class='error'>Agent not found</div>", status_code=400)
-
-    agent.new_conversation()
-    conv_id = agent.ensure_conversation()
-
-    convs = agent.db.list_conversations(limit=50, agent_name=name)
-    sidebar_html = _templates.get_template("partials/chat_sidebar.html").render(
-        conversations=convs
-    )
-
-    html = (
-        f'<div id="chat-messages"></div>'
-        f'<div id="sidebar-conversations" hx-swap-oob="innerHTML">'
-        f'{sidebar_html}</div>'
-    )
-
-    headers = {"HX-Push-Url": f"/?conv={conv_id}"}
-    return HTMLResponse(html, headers=headers)

cheddahbot/web/routes_pages.py

@@ -1,172 +0,0 @@
-"""Page routes: GET / (chat), GET /dashboard, dashboard partials."""
-
-from __future__ import annotations
-
-import logging
-from datetime import UTC, datetime
-from typing import TYPE_CHECKING
-
-from fastapi import APIRouter, Request
-from fastapi.responses import HTMLResponse
-from fastapi.templating import Jinja2Templates
-
-if TYPE_CHECKING:
-    from ..agent_registry import AgentRegistry
-    from ..config import Config
-    from ..db import Database
-    from ..llm import LLMAdapter
-    from ..scheduler import Scheduler
-
-log = logging.getLogger(__name__)
-
-router = APIRouter()
-
-_registry: AgentRegistry | None = None
-_config: Config | None = None
-_llm: LLMAdapter | None = None
-_db: Database | None = None
-_scheduler: Scheduler | None = None
-_templates: Jinja2Templates | None = None
-
-
-def setup(registry, config, llm, templates, db=None, scheduler=None):
-    global _registry, _config, _llm, _templates, _db, _scheduler
-    _registry = registry
-    _config = config
-    _llm = llm
-    _templates = templates
-    _db = db
-    _scheduler = scheduler
-
-
-@router.get("/")
-async def chat_page(request: Request):
-    agent_names = _registry.list_agents() if _registry else []
-    agents = []
-    for name in agent_names:
-        agent = _registry.get(name)
-        display = agent.agent_config.display_name if agent else name
-        agents.append({"name": name, "display_name": display})
-
-    default_agent = _registry.default_name if _registry else "default"
-    chat_model = _config.chat_model if _config else "unknown"
-    exec_available = _llm.is_execution_brain_available() if _llm else False
-    clickup_enabled = _config.clickup.enabled if _config else False
-
-    return _templates.TemplateResponse("chat.html", {
-        "request": request,
-        "agents": agents,
-        "default_agent": default_agent,
-        "chat_model": chat_model,
-        "exec_available": exec_available,
-        "clickup_enabled": clickup_enabled,
-    })
-
-
-@router.get("/dashboard")
-async def dashboard_page(request: Request):
-    return _templates.TemplateResponse("dashboard.html", {
-        "request": request,
-    })
-
-
-@router.get("/dashboard/pipeline")
-async def dashboard_pipeline():
-    """Return pipeline panel HTML partial with task data."""
-    if not _config or not _config.clickup.enabled:
-        return HTMLResponse('<p class="text-muted">ClickUp not configured</p>')
-
-    try:
-        from ..api import get_tasks
-        data = await get_tasks()
-        all_tasks = data.get("tasks", [])
-    except Exception as e:
-        log.error("Pipeline data fetch failed: %s", e)
-        return HTMLResponse(f'<p class="text-err">Error: {e}</p>')
-
-    # Group by work category, then by status
-    pipeline_statuses = [
-        "to do", "automation underway", "outline review", "internal review", "error",
-    ]
-    categories = {}  # category -> {status -> [tasks]}
-    for t in all_tasks:
-        cat = t.get("task_type") or "Other"
-        status = t.get("status", "unknown")
-
-        # Only show tasks in pipeline-relevant statuses
-        if status not in pipeline_statuses:
-            continue
-
-        if cat not in categories:
-            categories[cat] = {}
-        categories[cat].setdefault(status, []).append(t)
-
-    # Build HTML
-    html_parts = []
-
-    # Status summary counts
-    total_counts = {}
-    for cat_data in categories.values():
-        for status, tasks in cat_data.items():
-            total_counts[status] = total_counts.get(status, 0) + len(tasks)
-
-    if total_counts:
-        html_parts.append('<div class="pipeline-stats">')
-        for status in pipeline_statuses:
-            count = total_counts.get(status, 0)
-            html_parts.append(
-                f'<div class="pipeline-stat">'
-                f'<div class="stat-count">{count}</div>'
-                f'<div class="stat-label">{status}</div>'
-                f'</div>'
-            )
-        html_parts.append('</div>')
-
-    # Per-category tables
-    for cat_name in sorted(categories.keys()):
-        cat_data = categories[cat_name]
-        all_cat_tasks = []
-        for status in pipeline_statuses:
-            all_cat_tasks.extend(cat_data.get(status, []))
-
-        if not all_cat_tasks:
-            continue
-
-        html_parts.append(f'<div class="pipeline-group"><h4>{cat_name} ({len(all_cat_tasks)})</h4>')
-        html_parts.append('<table class="task-table"><thead><tr>'
-                         '<th>Task</th><th>Customer</th><th>Status</th><th>Due</th>'
-                         '</tr></thead><tbody>')
-
-        for task in all_cat_tasks:
-            name = task.get("name", "")
-            url = task.get("url", "")
-            customer = (task.get("custom_fields") or {}).get("Client", "N/A")
-            status = task.get("status", "")
-            status_class = "status-" + status.replace(" ", "-")
-
-            # Format due date
-            due_display = "-"
-            due_raw = task.get("due_date")
-            if due_raw:
-                try:
-                    due_dt = datetime.fromtimestamp(int(due_raw) / 1000, tz=UTC)
-                    due_display = due_dt.strftime("%b %d")
-                except (ValueError, TypeError, OSError):
-                    pass
-
-            name_cell = (
-                f'<a href="{url}" target="_blank">{name}</a>' if url else name
-            )
-
-            html_parts.append(
-                f'<tr><td>{name_cell}</td><td>{customer}</td>'
-                f'<td><span class="status-badge {status_class}">{status}</span></td>'
-                f'<td>{due_display}</td></tr>'
-            )
-
-        html_parts.append('</tbody></table></div>')
-
-    if not html_parts:
-        return HTMLResponse('<p class="text-muted">No active pipeline tasks</p>')
-
-    return HTMLResponse('\n'.join(html_parts))

cheddahbot/web/routes_sse.py

@@ -1,94 +0,0 @@
-"""SSE routes for live dashboard updates."""
-
-from __future__ import annotations
-
-import asyncio
-import json
-import logging
-from datetime import datetime
-from typing import TYPE_CHECKING
-
-from fastapi import APIRouter
-from sse_starlette.sse import EventSourceResponse
-
-if TYPE_CHECKING:
-    from ..db import Database
-    from ..notifications import NotificationBus
-    from ..scheduler import Scheduler
-
-log = logging.getLogger(__name__)
-
-router = APIRouter(prefix="/sse")
-
-_notification_bus: NotificationBus | None = None
-_scheduler: Scheduler | None = None
-_db: Database | None = None
-
-
-def setup(notification_bus, scheduler, db):
-    global _notification_bus, _scheduler, _db
-    _notification_bus = notification_bus
-    _scheduler = scheduler
-    _db = db
-
-
-@router.get("/notifications")
-async def sse_notifications():
-    """Stream new notifications as they arrive."""
-    listener_id = f"sse-notif-{id(asyncio.current_task())}"
-
-    # Subscribe to notification bus
-    queue: asyncio.Queue = asyncio.Queue()
-    loop = asyncio.get_event_loop()
-
-    if _notification_bus:
-        def on_notify(msg, cat):
-            loop.call_soon_threadsafe(
-                queue.put_nowait, {"message": msg, "category": cat}
-            )
-        _notification_bus.subscribe(listener_id, on_notify)
-
-    async def generate():
-        try:
-            while True:
-                try:
-                    notif = await asyncio.wait_for(queue.get(), timeout=30)
-                    yield {
-                        "event": "notification",
-                        "data": json.dumps(notif),
-                    }
-                except TimeoutError:
-                    yield {"event": "heartbeat", "data": ""}
-        finally:
-            if _notification_bus:
-                _notification_bus.unsubscribe(listener_id)
-
-    return EventSourceResponse(generate())
-
-
-@router.get("/loops")
-async def sse_loops():
-    """Push loop timestamps + active executions every 15s."""
-    async def generate():
-        while True:
-            data = {"loops": {}, "executions": {}}
-            if _scheduler:
-                ts = _scheduler.get_loop_timestamps()
-                data["loops"] = ts
-                # Serialize active executions (datetime -> str)
-                raw_exec = _scheduler.get_active_executions()
-                execs = {}
-                for tid, info in raw_exec.items():
-                    execs[tid] = {
-                        "name": info.get("name", ""),
-                        "tool": info.get("tool", ""),
-                        "started_at": info["started_at"].isoformat()
-                        if isinstance(info.get("started_at"), datetime)
-                        else str(info.get("started_at", "")),
-                        "thread": info.get("thread", ""),
-                    }
-                data["executions"] = execs
-            yield {"event": "loops", "data": json.dumps(data)}
-            await asyncio.sleep(15)
-
-    return EventSourceResponse(generate())

config.yaml

@@ -99,7 +99,7 @@ clickup:
 link_building:
   blm_dir: "E:/dev/Big-Link-Man"
   watch_folder: "//PennQnap1/SHARE1/cora-inbox"
-  watch_interval_minutes: 10
+  watch_interval_minutes: 60
   default_branded_plus_ratio: 0.7
 
 # AutoCora job submission

cora-link.md

@@ -1,287 +0,0 @@
-# Link Building Agent Plan
-
-## Context
-
-CheddahBot needs a link building agent that orchestrates the external Big-Link-Man CLI tool (`E:/dev/Big-Link-Man/`). The current workflow is manual: run Cora on another machine → get .xlsx → manually run `main.py ingest-cora` → manually run `main.py generate-batch`. This agent automates steps 2 and 3, triggered by folder watching, ClickUp tasks, or chat commands. It must be expandable for future link building methods (MCP server path, ingest-simple, etc.).
-
-## Decisions Made
-
-- **Watch folder**: `Z:/cora-inbox` (network drive, Cora machine accessible)
-- **File→task matching**: Fuzzy match .xlsx filename stem against ClickUp task's `Keyword` custom field
-- **New ClickUp field "LB Method"**: Dropdown with initial option "Cora Backlinks" (more added later)
-- **Dashboard**: API endpoint + NotificationBus events only (no frontend work — separate project)
-- **Sidecar files**: Not needed — all metadata comes from the matching ClickUp task
-- **Tool naming**: Orchestrator pattern — `run_link_building` is a thin dispatcher that reads `LB Method` and routes to the specific pipeline tool (e.g., `run_cora_backlinks`). Future link building methods get their own tools and slot into the orchestrator.
-
-## Files to Create
-
-### 1. `cheddahbot/tools/linkbuilding.py` — Main tool module
-
-Four `@tool`-decorated functions + private helpers:
-
-**`run_link_building(lb_method="", xlsx_path="", project_name="", money_site_url="", branded_plus_ratio=0.7, custom_anchors="", cli_flags="", ctx=None)`**
-- **Orchestrator/dispatcher** — reads `lb_method` (from ClickUp "LB Method" field or chat) and routes to the correct pipeline tool
-- If `lb_method` is "Cora Backlinks" or empty (default): calls `run_cora_backlinks()`
-- Future: if `lb_method` is "MCP Link Building": calls `run_mcp_link_building()` (not yet implemented)
-- Passes all other args through to the sub-tool
-- This is what the ClickUp skill_map always routes to
-
-**`run_cora_backlinks(xlsx_path, project_name, money_site_url, branded_plus_ratio=0.7, custom_anchors="", cli_flags="", ctx=None)`**
-- The actual Cora pipeline — runs ingest-cora → generate-batch
-- Step 1: Build CLI args, call `_run_blm_command(["ingest-cora", ...])`, parse stdout for job file path
-- Step 2: Call `_run_blm_command(["generate-batch", "-j", job_file, "--continue-on-error"])`
-- Updates KV store state and posts ClickUp comments at each step (following press_release.py pattern)
-- Returns `## ClickUp Sync` in output to signal scheduler that sync was handled internally
-- Can also be called directly from chat for explicit Cora runs
-
-**`blm_ingest_cora(xlsx_path, project_name, money_site_url, branded_plus_ratio=0.7, custom_anchors="", cli_flags="", ctx=None)`**
-- Standalone ingest — runs ingest-cora only, returns project ID and job file path
-- For cases where user wants to ingest but not generate yet
-
-**`blm_generate_batch(job_file, continue_on_error=True, debug=False, ctx=None)`**
-- Standalone generate — runs generate-batch only on an existing job file
-- For re-running generation or running a manually-created job
-
-**Private helpers:**
-- `_run_blm_command(args, timeout=1800)` — subprocess wrapper, runs `uv run python main.py <args>` from BLM_DIR, injects `-u`/`-p` from `BLM_USERNAME`/`BLM_PASSWORD` env vars
-- `_parse_ingest_output(stdout)` — regex extract project_id + job_file path
-- `_parse_generate_output(stdout)` — extract completion stats
-- `_build_ingest_args(...)` — construct CLI argument list from tool params
-- `_set_status(ctx, message)` — write pipeline status to KV store (for UI polling)
-- `_sync_clickup(ctx, task_id, step, message)` — post comment + update state
-
-**Critical: always pass `-m` flag** to ingest-cora to prevent interactive stdin prompt from blocking the subprocess.
-
-### 2. `skills/linkbuilding.md` — Skill file
-
-YAML frontmatter linking to `[run_link_building, run_cora_backlinks, blm_ingest_cora, blm_generate_batch, scan_cora_folder]` tools and `[link_builder, default]` agents. Markdown body describes when to use, default flags, workflow steps.
-
-### 3. `tests/test_linkbuilding.py` — Test suite (~40 tests)
-
-All tests mock `subprocess.run` — never call Big-Link-Man. Categories:
-- Output parser unit tests (`_parse_ingest_output`, `_parse_generate_output`)
-- CLI arg builder tests (all flag combinations, missing required params)
-- Full pipeline integration (happy path, ingest failure, generate failure)
-- ClickUp state machine (executing → completed, executing → failed)
-- Folder watcher scan logic (new files, skip processed, missing ClickUp match)
-
-## Files to Modify
-
-### 4. `cheddahbot/config.py` — Add LinkBuildingConfig
-
-```python
-@dataclass
-class LinkBuildingConfig:
-    blm_dir: str = "E:/dev/Big-Link-Man"
-    watch_folder: str = ""                    # empty = disabled
-    watch_interval_minutes: int = 60
-    default_branded_plus_ratio: float = 0.7
-```
-
-Add `link_building: LinkBuildingConfig` field to `Config` dataclass. Add YAML loading block in `load_config()` (same pattern as memory/scheduler/shell). Add env var override for `BLM_DIR`.
-
-### 5. `config.yaml` — Three additions
-
-**New top-level section:**
-```yaml
-link_building:
-  blm_dir: "E:/dev/Big-Link-Man"
-  watch_folder: "Z:/cora-inbox"
-  watch_interval_minutes: 60
-  default_branded_plus_ratio: 0.7
-```
-
-**New skill_map entry under clickup:**
-```yaml
-"Link Building":
-  tool: "run_link_building"
-  auto_execute: false           # Cora Backlinks triggered by folder watcher, not scheduler
-  complete_status: "complete"   # Override: use "complete" instead of "internal review"
-  error_status: "internal review"  # On failure, move to internal review
-  field_mapping:
-    lb_method: "LB Method"
-    project_name: "task_name"
-    money_site_url: "IMSURL"
-    custom_anchors: "CustomAnchors"
-    branded_plus_ratio: "BrandedPlusRatio"
-    cli_flags: "CLIFlags"
-    xlsx_path: "CoraFile"
-```
-
-**New agent:**
-```yaml
-- name: link_builder
-  display_name: Link Builder
-  tools: [run_link_building, run_cora_backlinks, blm_ingest_cora, blm_generate_batch, scan_cora_folder, delegate_task, remember, search_memory]
-  memory_scope: ""
-```
-
-### 6. `cheddahbot/scheduler.py` — Add folder watcher (4th daemon thread)
-
-**New thread `_folder_watch_loop`** alongside existing poll, heartbeat, and ClickUp threads:
-- Starts if `config.link_building.watch_folder` is non-empty
-- Runs every `watch_interval_minutes` (default 60)
-- `_scan_watch_folder()` globs `*.xlsx` in watch folder
-- For each file, checks KV store `linkbuilding:watched:{filename}` — skip if already processed
-- **Fuzzy-matches filename stem against ClickUp tasks** with `LB Method = "Cora Backlinks"` and status "to do":
-  - Queries ClickUp for Link Building tasks
-  - Compares normalized filename stem against each task's `Keyword` custom field
-  - If match found: extracts money_site_url from IMSURL field, cli_flags from CLIFlags field, etc.
-  - If no match: logs warning, marks as "unmatched" in KV store, sends notification asking user to create/link a ClickUp task
-- On match: executes `run_link_building` tool with args from the ClickUp task fields
-- On completion: moves .xlsx to `Z:/cora-inbox/processed/` subfolder, updates KV state
-- On failure: updates KV state with error, notifies via NotificationBus
-
-**File handling after pipeline:**
-- On success: .xlsx moved from `Z:/cora-inbox/` → `Z:/cora-inbox/processed/`
-- On failure: .xlsx stays in `Z:/cora-inbox/` (KV store marks it as failed so watcher doesn't retry automatically; user can reset KV entry to retry)
-
-**Also adds `scan_cora_folder` tool** (can live in linkbuilding.py):
-- Chat-invocable utility for the agent to check what's in the watch folder
-- Returns list of unprocessed .xlsx files with ClickUp match status
-- Internal agent tool, not a dashboard concern
-
-### 7. `cheddahbot/clickup.py` — Add field creation method
-
-Add `create_custom_field(list_id, name, field_type, type_config=None)` method that calls `POST /list/{list_id}/field`. Used by the setup tool to auto-create custom fields across lists.
-
-### 8. `cheddahbot/__main__.py` — Add API endpoint
-
-Add before Gradio mount:
-```python
-@fastapi_app.get("/api/linkbuilding/status")
-async def linkbuilding_status():
-    """Return link building status for dashboard consumption."""
-    # Returns:
-    # {
-    #   "pending_cora_runs": [
-    #     {"keyword": "precision cnc machining", "url": "https://...", "client": "Chapter 2", "task_id": "abc123"},
-    #     ...
-    #   ],
-    #   "in_progress": [...],     # Currently executing pipelines
-    #   "completed": [...],       # Recently completed (last 7 days)
-    #   "failed": [...]           # Failed tasks needing attention
-    # }
-```
-
-The `pending_cora_runs` section is the key dashboard data: queries ClickUp for "to do" tasks with Work Category="Link Building" and LB Method="Cora Backlinks", returns each task's `Keyword` field and `IMSURL` (copiable URL) so the user can see exactly which Cora reports need to be run.
-
-Also push link building events to NotificationBus (category="linkbuilding") at each pipeline step for future real-time dashboard support.
-
-No other `__main__.py` changes needed — agent wiring is automatic from config.yaml.
-
-## ClickUp Custom Fields (Auto-Created)
-
-New custom fields to be created programmatically:
-
-| Field | Type | Purpose |
-|-------|------|---------|
-| `LB Method` | Dropdown | Link building subtype. Initial option: "Cora Backlinks" |
-| `Keyword` | Short Text | Target keyword (used for file matching) |
-| `CoraFile` | Short Text | Path to .xlsx file (optional, set by agent after file match) |
-| `CustomAnchors` | Short Text | Comma-separated anchor text overrides |
-| `BrandedPlusRatio` | Short Text | Override for `-bp` flag (e.g., "0.7") |
-| `CLIFlags` | Short Text | Raw additional CLI flags (e.g., "-r 5 -t 0.3") |
-
-Fields that already exist and will be reused: `Client`, `IMSURL`, `Work Category` (add "Link Building" option).
-
-### Auto-creation approach
-
-- Add `create_custom_field(list_id, name, type, type_config=None)` method to `cheddahbot/clickup.py` — calls `POST /list/{list_id}/field`
-- Add a `setup_linkbuilding_fields` tool (category="linkbuilding") that:
-  1. Gets all list IDs in the space
-  2. For each list, checks if fields already exist (via `get_custom_fields`)
-  3. Creates missing fields via the new API method
-  4. For `LB Method` dropdown, creates with `type_config` containing "Cora Backlinks" option
-  5. For `Work Category`, adds "Link Building" option if missing
-- This tool runs once during initial setup, or can be re-run if new lists are added
-- Also add "Link Building" as an option to the existing `Work Category` dropdown if not present
-
-## Data Flow & Status Lifecycle
-
-### Primary Trigger: Folder Watcher (Cora Backlinks)
-
-The folder watcher is the main trigger for Cora Backlinks. The ClickUp scheduler does NOT auto-execute these — it can't, because the .xlsx doesn't exist until the user runs Cora.
-
-```
-1. ClickUp task created:
-   Work Category="Link Building", LB Method="Cora Backlinks", status="to do"
-   Fields filled: Client, IMSURL, Keyword, CLIFlags, BrandedPlusRatio, etc.
-   → Appears on dashboard as "needs Cora run"
-
-2. User runs Cora manually, drops .xlsx in Z:/cora-inbox
-
-3. Folder watcher (_scan_watch_folder, runs every 60 min):
-   → Finds precision-cnc-machining.xlsx
-   → Fuzzy matches "precision cnc machining" against Keyword field on ClickUp "to do" Link Building tasks
-   → Match found → extracts metadata from ClickUp task (IMSURL, CLIFlags, etc.)
-   → Sets CoraFile field on the ClickUp task to the file path
-   → Moves task to "in progress"
-   → Posts comment: "Starting Cora Backlinks pipeline..."
-
-4. Pipeline runs:
-   → Step 1: ingest-cora → comment: "CORA report ingested. Job file: jobs/xxx.json"
-   → Step 2: generate-batch → comment: "Content generation complete. X articles across Y tiers."
-
-5. On success:
-   → Move task to "complete"
-   → Post summary comment with stats
-   → Move .xlsx to Z:/cora-inbox/processed/
-
-6. On failure:
-   → Move task to "internal review"
-   → Post error comment with details
-   → .xlsx stays in Z:/cora-inbox (can retry)
-```
-
-### Secondary Trigger: Chat
-
-```
-User: "Run link building for Z:/cora-inbox/precision-cnc-machining.xlsx"
-  → Chat brain calls run_cora_backlinks (or run_link_building with explicit lb_method)
-  → Tool auto-looks up matching ClickUp task via Keyword field (if exists)
-  → Same pipeline + ClickUp sync as above
-  → If no ClickUp match: runs pipeline without ClickUp tracking, returns results to chat only
-```
-
-### Future Trigger: ClickUp Scheduler (other LB Methods)
-
-Future link building methods (MCP, etc.) that don't need a .xlsx CAN be auto-executed by the ClickUp scheduler. The `run_link_building` orchestrator checks `lb_method`:
-- "Cora Backlinks" → requires xlsx_path, skips if empty (folder watcher handles these)
-- Future methods → can execute directly from ClickUp task data
-
-### ClickUp Skill Map Note
-
-The skill_map entry for "Link Building" exists primarily for **field mapping reference** (so the folder watcher and chat know which ClickUp fields map to which tool params). The ClickUp scheduler will discover these tasks but `run_link_building` will skip Cora Backlinks that have no xlsx_path — they're waiting for the folder watcher.
-
-## Implementation Order
-
-1. **Config** — Add `LinkBuildingConfig` to config.py, add `link_building:` section to config.yaml, add `link_builder` agent to config.yaml
-2. **Core tools** — Create `cheddahbot/tools/linkbuilding.py` with `_run_blm_command`, parsers, `run_link_building` orchestrator, and `run_cora_backlinks` pipeline
-3. **Standalone tools** — Add `blm_ingest_cora` and `blm_generate_batch`
-4. **Tests** — Create `tests/test_linkbuilding.py`, verify with `uv run pytest tests/test_linkbuilding.py -v`
-5. **ClickUp field creation** — Add `create_custom_field` to clickup.py, add `setup_linkbuilding_fields` tool
-6. **ClickUp integration** — Add skill_map entry, add ClickUp state tracking to tools
-7. **Folder watcher** — Add `_folder_watch_loop` to scheduler.py, add `scan_cora_folder` tool
-8. **API endpoint** — Add `/api/linkbuilding/status` to `__main__.py`
-9. **Skill file** — Create `skills/linkbuilding.md`
-10. **ClickUp setup** — Run `setup_linkbuilding_fields` to auto-create custom fields across all lists
-11. **Full test run** — `uv run pytest -v --no-cov`
-
-## Verification
-
-1. **Unit tests**: `uv run pytest tests/test_linkbuilding.py -v` — all pass with mocked subprocess
-2. **Full suite**: `uv run pytest -v --no-cov` — no regressions
-3. **Lint**: `uv run ruff check .` + `uv run ruff format .`
-4. **Manual e2e**: Drop a real .xlsx in Z:/cora-inbox, verify ingest-cora runs, job JSON created, generate-batch runs
-5. **ClickUp e2e**: Create a Link Building task in ClickUp with proper fields, wait for scheduler poll, verify execution
-6. **Chat e2e**: Ask CheddahBot to "run link building for [keyword]" via chat UI
-7. **API check**: Hit `http://localhost:7860/api/linkbuilding/status` and verify data returned
-
-## Key Reference Files
-
-- `cheddahbot/tools/press_release.py` — Reference pattern for multi-step pipeline tool
-- `cheddahbot/scheduler.py:55-76` — Where to add 4th daemon thread
-- `cheddahbot/config.py:108-200` — load_config() pattern for new config sections
-- `E:/dev/Big-Link-Man/docs/CLI_COMMAND_REFERENCE.md` — Full CLI reference
-- `E:/dev/Big-Link-Man/src/cli/commands.py` — Exact output formats to parse

docs/ARCHITECTURE.md

@@ -1,721 +0,0 @@
-# CheddahBot Architecture
-
-## System Overview
-
-CheddahBot is a personal AI assistant built in Python. It exposes a Gradio-based
-web UI, routes user messages through an agent loop backed by a model-agnostic LLM
-adapter, persists conversations in SQLite, maintains a 4-layer memory system with
-optional semantic search, and provides an extensible tool registry that the LLM
-can invoke mid-conversation. A background scheduler handles cron-based tasks and
-periodic heartbeat checks.
-
-### Data Flow Diagram
-
-```
-User (browser)
-     |
-     v
-+-----------+      +------------+      +--------------+
-| Gradio UI | ---> |   Agent    | ---> |  LLM Adapter |
-|  (ui.py)  |      | (agent.py) |      |   (llm.py)   |
-+-----------+      +-----+------+      +------+-------+
-                         |                     |
-            +------------+-------+     +-------+--------+
-            |            |       |     | Claude CLI     |
-            v            v       v     | OpenRouter     |
-      +---------+  +---------+ +---+  | Ollama         |
-      | Router  |  | Tools   | | DB|  | LM Studio      |
-      |(router) |  |(tools/) | |(db|  +----------------+
-      +----+----+  +----+----+ +---+
-           |            |
-   +-------+--+    +----+----+
-   | Identity |    | Memory  |
-   | SOUL.md  |    | System  |
-   | USER.md  |    |(memory) |
-   +----------+    +---------+
-```
-
-1. The user submits text (or voice / files) through the Gradio interface.
-2. `ui.py` hands the message to `Agent.respond()`.
-3. The agent stores the user message in SQLite, builds a system prompt via
-   `router.py` (loading identity files and memory context), and formats the
-   conversation history.
-4. The agent sends messages to `LLMAdapter.chat()` which dispatches to the
-   correct provider backend.
-5. The LLM response streams back. If it contains tool-call requests, the agent
-   executes them through `ToolRegistry.execute()`, appends the results, and loops
-   back to step 4 (up to 10 iterations).
-6. The final assistant response is stored in the database and streamed to the UI.
-7. After responding, the agent checks whether the conversation has exceeded the
-   flush threshold; if so, the memory system summarizes older messages into the
-   daily log.
-
----
-
-## Module-by-Module Breakdown
-
-### `__main__.py` -- Entry Point
-
-**File:** `cheddahbot/__main__.py`
-
-Orchestrates startup in this order:
-
-1. `load_config()` -- loads configuration from env vars / YAML / defaults.
-2. `Database(config.db_path)` -- opens (or creates) the SQLite database.
-3. `LLMAdapter(...)` -- initializes the model-agnostic LLM client.
-4. `Agent(config, db, llm)` -- creates the core agent.
-5. `MemorySystem(config, db)` -- initializes the memory system and injects it
-   into the agent via `agent.set_memory()`.
-6. `ToolRegistry(config, db, agent)` -- auto-discovers and loads all tool
-   modules, then injects via `agent.set_tools()`.
-7. `Scheduler(config, db, agent)` -- starts two daemon threads (task poller and
-   heartbeat).
-8. `create_ui(agent, config, llm)` -- builds the Gradio Blocks app and launches
-   it on the configured host/port.
-
-Each subsystem (memory, tools, scheduler) is wrapped in a try/except so the
-application degrades gracefully if optional dependencies are missing.
-
----
-
-### `config.py` -- Configuration
-
-**File:** `cheddahbot/config.py`
-
-Defines four dataclasses:
-
-| Dataclass        | Key Fields                                                    |
-|------------------|---------------------------------------------------------------|
-| `Config`         | `default_model`, `host`, `port`, `ollama_url`, `lmstudio_url`, `openrouter_api_key`, plus derived paths (`root_dir`, `data_dir`, `identity_dir`, `memory_dir`, `skills_dir`, `db_path`) |
-| `MemoryConfig`   | `max_context_messages` (50), `flush_threshold` (40), `embedding_model` ("all-MiniLM-L6-v2"), `search_top_k` (5) |
-| `SchedulerConfig` | `heartbeat_interval_minutes` (30), `poll_interval_seconds` (60) |
-| `ShellConfig`    | `blocked_commands`, `require_approval` (False)                |
-
-`load_config()` applies three layers of configuration in priority order:
-
-1. Dataclass defaults (lowest priority).
-2. `config.yaml` at the project root (middle priority).
-3. Environment variables with the `CHEDDAH_` prefix, plus `OPENROUTER_API_KEY`
-   (highest priority).
-
-The function also ensures required data directories exist on disk.
-
----
-
-### `db.py` -- Database Layer
-
-**File:** `cheddahbot/db.py`
-
-A thin wrapper around SQLite using thread-local connections (one connection per
-thread), WAL journal mode, and foreign keys.
-
-**Key methods:**
-
-- `create_conversation(conv_id, title)` -- insert a new conversation row.
-- `list_conversations(limit)` -- return recent conversations ordered by
-  `updated_at`.
-- `add_message(conv_id, role, content, ...)` -- insert a message and touch the
-  conversation's `updated_at`.
-- `get_messages(conv_id, limit)` -- return messages in chronological order.
-- `count_messages(conv_id)` -- count messages for flush-threshold checks.
-- `add_scheduled_task(name, prompt, schedule)` -- persist a scheduled task.
-- `get_due_tasks()` -- return tasks whose `next_run` is in the past or NULL.
-- `update_task_next_run(task_id, next_run)` -- update the next execution time.
-- `log_task_run(task_id, result, error)` -- record the outcome of a task run.
-- `kv_set(key, value)` / `kv_get(key)` -- generic key-value store.
-
----
-
-### `agent.py` -- Core Agent Loop
-
-**File:** `cheddahbot/agent.py`
-
-Contains the `Agent` class, the central coordinator.
-
-**Key members:**
-
-- `conv_id` -- current conversation ID (a 12-character hex string).
-- `_memory` -- optional `MemorySystem` reference.
-- `_tools` -- optional `ToolRegistry` reference.
-
-**Primary method: `respond(user_input, files)`**
-
-This is a Python generator that yields text chunks for streaming. The detailed
-flow is described in the next section.
-
-**Helper: `respond_to_prompt(prompt)`**
-
-Non-streaming wrapper that collects all chunks and returns a single string. Used
-by the scheduler and heartbeat for internal prompts.
-
----
-
-### `router.py` -- System Prompt Builder
-
-**File:** `cheddahbot/router.py`
-
-Two functions:
-
-1. `build_system_prompt(identity_dir, memory_context, tools_description)` --
-   assembles the full system prompt by concatenating these sections separated by
-   horizontal rules:
-   - Contents of `identity/SOUL.md`
-   - Contents of `identity/USER.md`
-   - Memory context string (from the memory system)
-   - Tools description listing (from the tool registry)
-   - A fixed "Instructions" section with core behavioral directives.
-
-2. `format_messages_for_llm(system_prompt, history, max_messages)` --
-   converts raw database rows into the `[{role, content}]` format expected by
-   the LLM. The system prompt becomes the first message. Tool results are
-   converted to user messages prefixed with `[Tool Result]`. History is trimmed
-   to the most recent `max_messages` entries.
-
----
-
-### `llm.py` -- LLM Adapter
-
-**File:** `cheddahbot/llm.py`
-
-Described in detail in a dedicated section below.
-
----
-
-### `memory.py` -- Memory System
-
-**File:** `cheddahbot/memory.py`
-
-Described in detail in a dedicated section below.
-
----
-
-### `media.py` -- Audio/Video Processing
-
-**File:** `cheddahbot/media.py`
-
-Three utility functions:
-
-- `transcribe_audio(path)` -- Speech-to-text. Tries local Whisper first, then
-  falls back to the OpenAI Whisper API.
-- `text_to_speech(text, output_path, voice)` -- Text-to-speech via `edge-tts`
-  (free, no API key). Defaults to the `en-US-AriaNeural` voice.
-- `extract_video_frames(video_path, max_frames)` -- Extracts key frames from
-  video using `ffprobe` (to get duration) and `ffmpeg` (to extract JPEG frames).
-
----
-
-### `scheduler.py` -- Scheduler and Heartbeat
-
-**File:** `cheddahbot/scheduler.py`
-
-Described in detail in a dedicated section below.
-
----
-
-### `ui.py` -- Gradio Web Interface
-
-**File:** `cheddahbot/ui.py`
-
-Builds a Gradio Blocks application with:
-
-- A model dropdown (populated from `llm.list_available_models()`) with a refresh
-  button and a "New Chat" button.
-- A `gr.Chatbot` widget for the conversation (500px height, copy buttons).
-- A `gr.MultimodalTextbox` supporting text, file upload, and microphone input.
-- A "Voice Chat" accordion for record-and-respond audio interaction.
-- A "Conversation History" accordion showing past conversations from the
-  database.
-- A "Settings" accordion with guidance on editing identity and config files.
-
-**Event wiring:**
-
-- Model dropdown change calls `llm.switch_model()`.
-- Refresh button re-discovers local models.
-- Message submit calls `agent.respond()` in streaming mode, updating the chatbot
-  widget with each chunk.
-- Audio files attached to messages are transcribed via `media.transcribe_audio()`
-  before being sent to the agent.
-- Voice Chat records audio, transcribes it, gets a text response from the agent,
-  converts it to speech via `media.text_to_speech()`, and plays it back.
-
----
-
-### `tools/__init__.py` -- Tool Registry
-
-**File:** `cheddahbot/tools/__init__.py`
-
-Described in detail in a dedicated section below.
-
----
-
-### `skills/__init__.py` -- Skill Registry
-
-**File:** `cheddahbot/skills/__init__.py`
-
-Defines a parallel registry for "skills" (multi-step operations). Key pieces:
-
-- `SkillDef` -- dataclass holding `name`, `description`, `func`.
-- `@skill(name, description)` -- decorator that registers a skill in the global
-  `_SKILLS` dict.
-- `load_skill(path)` -- dynamically loads a `.py` file as a module (triggering
-  any `@skill` decorators inside it).
-- `discover_skills(skills_dir)` -- loads all `.py` files from the skills
-  directory.
-- `list_skills()` / `run_skill(name, **kwargs)` -- query and execute skills.
-
----
-
-### `providers/__init__.py` -- Provider Extensions
-
-**File:** `cheddahbot/providers/__init__.py`
-
-Reserved for future custom provider implementations. Currently empty.
-
----
-
-## The Agent Loop in Detail
-
-When `Agent.respond(user_input)` is called, the following sequence occurs:
-
-```
-1. ensure_conversation()
-   |-- Creates a new conversation in the DB if one doesn't exist
-   |
-2. db.add_message(conv_id, "user", user_input)
-   |-- Persists the user's message
-   |
-3. Build system prompt
-   |-- memory.get_context(user_input)  --> memory context string
-   |-- tools.get_tools_schema()        --> OpenAI-format JSON schemas
-   |-- tools.get_tools_description()   --> human-readable tool list
-   |-- router.build_system_prompt(identity_dir, memory_context, tools_description)
-   |
-4. Load conversation history from DB
-   |-- db.get_messages(conv_id, limit=max_context_messages)
-   |-- router.format_messages_for_llm(system_prompt, history, max_messages)
-   |
-5. AGENT LOOP (up to MAX_TOOL_ITERATIONS = 10):
-   |
-   |-- llm.chat(messages, tools=tools_schema, stream=True)
-   |     |-- Yields {"type":"text","content":"..."} chunks --> streamed to user
-   |     |-- Yields {"type":"tool_use","name":"...","input":{...}} chunks
-   |
-   |-- If no tool_calls: store assistant message, BREAK
-   |
-   |-- If tool_calls present:
-   |     |-- Store assistant message with tool_calls metadata
-   |     |-- For each tool call:
-   |     |     |-- yield "Using tool: <name>" indicator
-   |     |     |-- tools.execute(name, input) --> result string
-   |     |     |-- yield tool result (truncated to 2000 chars)
-   |     |     |-- db.add_message(conv_id, "tool", result)
-   |     |     |-- Append result to messages as user message
-   |     |-- Continue loop (LLM sees tool results and can respond or call more tools)
-   |
-6. After loop: check if memory flush is needed
-   |-- If message count > flush_threshold:
-   |     |-- memory.auto_flush(conv_id)
-```
-
-The loop allows the LLM to chain up to 10 consecutive tool calls before being
-cut off. Each tool result is injected back into the conversation as a user
-message so the LLM can reason about it in the next iteration.
-
----
-
-## LLM Adapter Design
-
-**File:** `cheddahbot/llm.py`
-
-### Provider Routing
-
-The `LLMAdapter` supports four provider paths. The active provider is determined
-by examining the current model ID:
-
-| Model ID Pattern            | Provider      | Backend                          |
-|-----------------------------|---------------|----------------------------------|
-| `claude-*`                  | `claude`      | Claude Code CLI (subprocess)     |
-| `local/ollama/<model>`      | `ollama`      | Ollama HTTP API (OpenAI-compat)  |
-| `local/lmstudio/<model>`    | `lmstudio`    | LM Studio HTTP API (OpenAI-compat) |
-| Anything else               | `openrouter`  | OpenRouter API (OpenAI-compat)   |
-
-### The `chat()` Method
-
-This is the single entry point. It accepts a list of messages, an optional tools
-schema, and a stream flag. It returns a generator yielding dictionaries:
-
-- `{"type": "text", "content": "..."}` -- a text chunk to display.
-- `{"type": "tool_use", "id": "...", "name": "...", "input": {...}}` -- a tool
-  invocation request.
-
-### Claude Code CLI Path (`_chat_claude_sdk`)
-
-For Claude models, CheddahBot shells out to the `claude` CLI binary (the Claude
-Code SDK):
-
-1. Separates system prompt, conversation history, and the latest user message
-   from the messages list.
-2. Builds a full system prompt by appending conversation history under a
-   "Conversation So Far" heading.
-3. Invokes `claude -p <prompt> --model <model> --output-format json --system-prompt <system>`.
-4. The `CLAUDECODE` environment variable is stripped from the subprocess
-   environment to avoid nested-session errors.
-5. Parses the JSON output and yields the `result` field as a text chunk.
-6. On Windows, `shell=True` is used for compatibility with npm-installed
-   binaries.
-
-### OpenAI-Compatible Path (`_chat_openai_sdk`)
-
-For OpenRouter, Ollama, and LM Studio, the adapter uses the `openai` Python SDK:
-
-1. `_resolve_endpoint(provider)` returns the base URL and API key:
-   - OpenRouter: `https://openrouter.ai/api/v1` with the configured API key.
-   - Ollama: `http://localhost:11434/v1` with dummy key `"ollama"`.
-   - LM Studio: `http://localhost:1234/v1` with dummy key `"lm-studio"`.
-2. `_resolve_model_id(provider)` strips the `local/ollama/` or
-   `local/lmstudio/` prefix from the model ID.
-3. Creates an `openai.OpenAI` client with the resolved base URL and API key.
-4. In streaming mode: iterates over `client.chat.completions.create(stream=True)`,
-   accumulates tool call arguments across chunks (indexed by `tc.index`), yields
-   text deltas immediately, and yields completed tool calls at the end of the
-   stream.
-5. In non-streaming mode: makes a single call and yields text and tool calls from
-   the response.
-
-### Model Discovery
-
-- `discover_local_models()` -- probes the Ollama tags endpoint and LM Studio
-  models endpoint (3-second timeout each) and returns `ModelInfo` objects.
-- `list_available_models()` -- returns a combined list of hardcoded Claude
-  models, hardcoded OpenRouter models (if an API key is configured), and
-  dynamically discovered local models.
-
-### Model Switching
-
-`switch_model(model_id)` updates `current_model`. The `provider` property
-re-evaluates on every access, so switching models also implicitly switches
-providers.
-
----
-
-## Memory System
-
-**File:** `cheddahbot/memory.py`
-
-### The 4 Layers
-
-```
-Layer 1: Identity      -- identity/SOUL.md, identity/USER.md
-                          (loaded by router.py into the system prompt)
-
-Layer 2: Long-term     -- memory/MEMORY.md
-                          (persisted facts and instructions, appended over time)
-
-Layer 3: Daily logs    -- memory/YYYY-MM-DD.md
-                          (timestamped entries per day, including auto-flush summaries)
-
-Layer 4: Semantic      -- memory/embeddings.db
-                          (SQLite with vector embeddings for similarity search)
-```
-
-### How Memory Context is Built
-
-`MemorySystem.get_context(query)` is called once per agent turn. It assembles a
-string from:
-
-1. **Long-term memory** -- the last 2000 characters of `MEMORY.md`.
-2. **Today's log** -- the last 1500 characters of today's date file.
-3. **Semantic search results** -- the top-k most similar entries to the user's
-   query, formatted as a bulleted list.
-
-This string is injected into the system prompt by `router.py` under the heading
-"Relevant Memory".
-
-### Embedding and Search
-
-- The embedding model is `all-MiniLM-L6-v2` from `sentence-transformers` (lazy
-  loaded, thread-safe via a lock).
-- `_index_text(text, doc_id)` -- encodes the text into a vector and stores it in
-  `memory/embeddings.db` (table: `embeddings` with columns `id TEXT`, `text TEXT`,
-  `vector BLOB`).
-- `search(query, top_k)` -- encodes the query, loads all vectors from the
-  database, computes cosine similarity against each one, sorts by score, and
-  returns the top-k results.
-- If `sentence-transformers` is not installed, `_fallback_search()` performs
-  simple case-insensitive substring matching across all `.md` files in the memory
-  directory.
-
-### Writing to Memory
-
-- `remember(text)` -- appends a timestamped entry to `memory/MEMORY.md` and
-  indexes it for semantic search. Exposed to the LLM via the `remember_this`
-  tool.
-- `log_daily(text)` -- appends a timestamped entry to today's daily log file and
-  indexes it. Exposed via the `log_note` tool.
-
-### Auto-Flush
-
-When `Agent.respond()` finishes, it checks `db.count_messages(conv_id)`. If the
-count exceeds `config.memory.flush_threshold` (default 40):
-
-1. `auto_flush(conv_id)` loads up to 200 messages.
-2. All but the last 10 are selected for summarization.
-3. A summary string is built from the selected messages (truncated to 1000
-   chars).
-4. The summary is appended to the daily log via `log_daily()`.
-
-This prevents conversations from growing unbounded while preserving context in
-the daily log for future semantic search.
-
-### Reindexing
-
-`reindex_all()` clears all embeddings and re-indexes every line (longer than 10
-characters) from every `.md` file in the memory directory. This can be called
-to rebuild the search index from scratch.
-
----
-
-## Tool System
-
-**File:** `cheddahbot/tools/__init__.py` (registry) and `cheddahbot/tools/*.py`
-(tool modules)
-
-### The `@tool` Decorator
-
-```python
-from cheddahbot.tools import tool
-
-@tool("my_tool_name", "Description of what this tool does", category="general")
-def my_tool_name(param1: str, param2: int = 10) -> str:
-    return f"Result: {param1}, {param2}"
-```
-
-The decorator:
-
-1. Creates a `ToolDef` object containing the function, name, description,
-   category, and auto-extracted parameter schema.
-2. Registers it in the global `_TOOLS` dictionary keyed by name.
-3. Attaches the `ToolDef` as `func._tool_def` on the original function.
-
-### Parameter Schema Generation
-
-`_extract_params(func)` inspects the function signature using `inspect`:
-
-- Skips parameters named `self` or `ctx`.
-- Maps type annotations to JSON Schema types: `str` -> `"string"`, `int` ->
-  `"integer"`, `float` -> `"number"`, `bool` -> `"boolean"`, `list` ->
-  `"array"`. Unannotated parameters default to `"string"`.
-- Parameters without defaults are marked as required.
-
-### Schema Output
-
-`ToolDef.to_openai_schema()` returns the tool definition in OpenAI
-function-calling format:
-
-```json
-{
-  "type": "function",
-  "function": {
-    "name": "tool_name",
-    "description": "...",
-    "parameters": {
-      "type": "object",
-      "properties": { ... },
-      "required": [ ... ]
-    }
-  }
-}
-```
-
-### Auto-Discovery
-
-When `ToolRegistry.__init__()` is called, `_discover_tools()` uses
-`pkgutil.iter_modules` to find every `.py` file in `cheddahbot/tools/` (skipping
-files starting with `_`). Each module is imported via `importlib.import_module`,
-which triggers the `@tool` decorators and populates the global registry.
-
-### Tool Execution
-
-`ToolRegistry.execute(name, args)`:
-
-1. Looks up the `ToolDef` in the global `_TOOLS` dict.
-2. Inspects the function signature for a `ctx` parameter. If present, injects a
-   context dictionary containing `config`, `db`, `agent`, and `memory`.
-3. Calls the function with the provided arguments.
-4. Returns the result as a string (or `"Done."` if the function returns `None`).
-5. Catches all exceptions and returns `"Tool error: ..."`.
-
-### Meta-Tools
-
-Two special tools enable runtime extensibility:
-
-**`build_tool`** (in `cheddahbot/tools/build_tool.py`):
-- Accepts `name`, `description`, and `code` (Python source using the `@tool`
-  decorator).
-- Writes a new `.py` file into `cheddahbot/tools/`.
-- Hot-imports the module via `importlib.import_module`, which triggers the
-  `@tool` decorator and registers the new tool immediately.
-- If the import fails, the file is deleted.
-
-**`build_skill`** (in `cheddahbot/tools/build_skill.py`):
-- Accepts `name`, `description`, and `steps` (Python source using the `@skill`
-  decorator).
-- Writes a new `.py` file into the configured `skills/` directory.
-- Calls `skills.load_skill()` to dynamically import it.
-
----
-
-## Scheduler and Heartbeat Design
-
-**File:** `cheddahbot/scheduler.py`
-
-The `Scheduler` class starts two daemon threads at application boot.
-
-### Task Poller Thread
-
-- Runs in `_poll_loop()`, sleeping for `poll_interval_seconds` (default 60)
-  between iterations.
-- Each iteration calls `_run_due_tasks()`:
-  1. Queries `db.get_due_tasks()` for tasks where `next_run` is NULL or in the
-     past.
-  2. For each due task, calls `agent.respond_to_prompt(task["prompt"])` to
-     generate a response.
-  3. Logs the result via `db.log_task_run()`.
-  4. If the schedule is `"once:<datetime>"`, the task is disabled.
-  5. Otherwise, the schedule is treated as a cron expression: `croniter` is used
-     to calculate the next run time, which is saved via
-     `db.update_task_next_run()`.
-
-### Heartbeat Thread
-
-- Runs in `_heartbeat_loop()`, sleeping for `heartbeat_interval_minutes`
-  (default 30) between iterations.
-- Waits 60 seconds before the first heartbeat to let the system initialize.
-- Each iteration calls `_run_heartbeat()`:
-  1. Reads `identity/HEARTBEAT.md`.
-  2. Sends the checklist to the agent as a prompt: "HEARTBEAT CHECK. Review this
-     checklist and take action if needed."
-  3. If the response contains `"HEARTBEAT_OK"`, no action is logged.
-  4. Otherwise, the response is logged to the daily log via
-     `memory.log_daily()`.
-
-### Thread Safety
-
-Both threads are daemon threads (they die when the main process exits). The
-`_stop_event` threading event can be set to gracefully shut down both loops. The
-database layer uses thread-local connections, so concurrent access from the
-scheduler threads and the Gradio request threads is safe.
-
----
-
-## Database Schema
-
-The SQLite database (`data/cheddahbot.db`) contains five tables:
-
-### `conversations`
-
-| Column       | Type | Notes              |
-|--------------|------|--------------------|
-| `id`         | TEXT | Primary key (hex)  |
-| `title`      | TEXT | Display title      |
-| `created_at` | TEXT | ISO 8601 UTC       |
-| `updated_at` | TEXT | ISO 8601 UTC       |
-
-### `messages`
-
-| Column        | Type    | Notes                                      |
-|---------------|---------|--------------------------------------------|
-| `id`          | INTEGER | Autoincrement primary key                  |
-| `conv_id`     | TEXT    | Foreign key to `conversations.id`          |
-| `role`        | TEXT    | `"user"`, `"assistant"`, or `"tool"`       |
-| `content`     | TEXT    | Message body                               |
-| `tool_calls`  | TEXT    | JSON array of `{name, input}` (nullable)   |
-| `tool_result` | TEXT    | Name of the tool that produced this result (nullable) |
-| `model`       | TEXT    | Model ID used for this response (nullable) |
-| `created_at`  | TEXT    | ISO 8601 UTC                               |
-
-Index: `idx_messages_conv` on `(conv_id, created_at)`.
-
-### `scheduled_tasks`
-
-| Column       | Type    | Notes                                 |
-|--------------|---------|---------------------------------------|
-| `id`         | INTEGER | Autoincrement primary key             |
-| `name`       | TEXT    | Human-readable task name              |
-| `prompt`     | TEXT    | The prompt to send to the agent       |
-| `schedule`   | TEXT    | Cron expression or `"once:<datetime>"`|
-| `enabled`    | INTEGER | 1 = active, 0 = disabled              |
-| `next_run`   | TEXT    | ISO 8601 UTC (nullable)               |
-| `created_at` | TEXT    | ISO 8601 UTC                          |
-
-### `task_run_logs`
-
-| Column        | Type    | Notes                              |
-|---------------|---------|------------------------------------|
-| `id`          | INTEGER | Autoincrement primary key          |
-| `task_id`     | INTEGER | Foreign key to `scheduled_tasks.id`|
-| `started_at`  | TEXT    | ISO 8601 UTC                       |
-| `finished_at` | TEXT    | ISO 8601 UTC (nullable)            |
-| `result`      | TEXT    | Agent response (nullable)          |
-| `error`       | TEXT    | Error message if failed (nullable) |
-
-### `kv_store`
-
-| Column  | Type | Notes           |
-|---------|------|-----------------|
-| `key`   | TEXT | Primary key     |
-| `value` | TEXT | Arbitrary value |
-
-### Embeddings Database
-
-A separate SQLite file at `memory/embeddings.db` holds one table:
-
-### `embeddings`
-
-| Column   | Type | Notes                                |
-|----------|------|--------------------------------------|
-| `id`     | TEXT | Primary key (e.g. `"daily:2026-02-14:08:30"`) |
-| `text`   | TEXT | The original text that was embedded  |
-| `vector` | BLOB | Raw float32 bytes of the embedding vector |
-
----
-
-## Identity Files
-
-Three Markdown files in the `identity/` directory define the agent's personality,
-user context, and background behavior.
-
-### `identity/SOUL.md`
-
-Defines the agent's personality, communication style, boundaries, and quirks.
-This is loaded first into the system prompt, making it the most prominent
-identity influence on every response.
-
-Contents are read by `router.build_system_prompt()` at the beginning of each
-agent turn.
-
-### `identity/USER.md`
-
-Contains a user profile template: name, technical level, primary language,
-current projects, and communication preferences. The user edits this file to
-customize how the agent addresses them and what context it assumes.
-
-Loaded by `router.build_system_prompt()` immediately after SOUL.md.
-
-### `identity/HEARTBEAT.md`
-
-A checklist of items to review on each heartbeat cycle. The scheduler reads this
-file and sends it to the agent as a prompt every `heartbeat_interval_minutes`
-(default 30 minutes). The agent processes the checklist and either confirms
-"HEARTBEAT_OK" or takes action and logs it.
-
-### Loading Order in the System Prompt
-
-The system prompt assembled by `router.build_system_prompt()` concatenates these
-sections, separated by `\n\n---\n\n`:
-
-1. SOUL.md contents
-2. USER.md contents
-3. Memory context (long-term + daily log + semantic search results)
-4. Tools description (categorized list of available tools)
-5. Core instructions (hardcoded behavioral directives)

docs/clickup-task-creation.md

@@ -1,61 +0,0 @@
-# ClickUp Task Creation
-
-## CLI Script
-
-```bash
-uv run python scripts/create_clickup_task.py --name "LINKS - keyword" --client "Client Name" \
-    --category "Link Building" --due-date 2026-03-18 --tag mar26 --time-estimate 2h \
-    --field "Keyword=keyword" --field "IMSURL=https://example.com" --field "LB Method=Cora Backlinks"
-```
-
-## Defaults
-
-- Priority: High (2)
-- Assignee: Bryan (10765627)
-- Status: "to do"
-- Due date format: YYYY-MM-DD
-- Tag format: mmmYY (e.g. feb26, mar26)
-
-## Custom Fields
-
-Any field can be set via `--field "Name=Value"`. Dropdowns are auto-resolved by name (case-insensitive).
-
-## Task Types
-
-### Link Building
-- **Prefix**: `LINKS - {keyword}`
-- **Work Category**: "Link Building"
-- **Required fields**: Keyword, IMSURL
-- **LB Method**: default "Cora Backlinks"
-- **CLIFlags**: only add `--tier1-count N` when count is specified
-- **BrandedPlusRatio**: default to 0.7
-- **CustomAnchors**: only if given a list of custom anchors
-- **time estimate**: 2.5h
-
-### On Page Optimization
-- **Prefix**: `OPT - {keyword}`
-- **Work Category**: "On Page Optimization"
-- **Required fields**: Keyword, IMSURL
-- **time estimate**: 3h
-- 
-
-### Content Creation
-- **Prefix**: `CREATE - {keyword}`
-- **Work Category**: "Content Creation"
-- **Required fields**: Keyword
-- **time estimate**: 4h
-
-### Press Release
-- **Prefix**: `PR - {keyword}`
-- **Required fields**: Keyword, IMSURL
-- **Work Category**: "Press Release"
-- **PR Topic**: if not provided, ask if there is a topic. it can be blank if they respond with none.
-- **time estimate**: 1.5h
-
-## Chat Tool
-
-The `clickup_create_task` tool provides the same capabilities via CheddahBot UI. Arbitrary custom fields are passed as JSON via `custom_fields_json`.
-
-## Client Folder Lookup
-
-Tasks are created in the "Overall" list inside the client's folder. Folder name is matched case-insensitively.

docs/ntfy-setup.md

@@ -1,110 +0,0 @@
-# ntfy.sh Push Notifications Setup
-
-CheddahBot sends push notifications to your phone and desktop via [ntfy.sh](https://ntfy.sh) when tasks complete, reports are ready, or errors occur.
-
-## 1. Install the ntfy App
-
-- **Android:** [Play Store](https://play.google.com/store/apps/details?id=io.heckel.ntfy)
-- **iOS:** [App Store](https://apps.apple.com/us/app/ntfy/id1625396347)
-- **Desktop:** Open [ntfy.sh](https://ntfy.sh) in your browser and enable browser notifications when prompted
-
-## 2. Pick Topic Names
-
-Topics are like channels. Anyone who knows the topic name can subscribe, so use random strings:
-
-```
-cheddahbot-a8f3k9x2m7
-cheddahbot-errors-p4w2j6n8
-```
-
-Generate your own — any random string works. No account or registration needed.
-
-## 3. Subscribe to Your Topics
-
-**Phone app:**
-1. Open the ntfy app
-2. Tap the + button
-3. Enter your topic name (e.g. `cheddahbot-a8f3k9x2m7`)
-4. Server: `https://ntfy.sh` (default)
-5. Repeat for your errors topic
-
-**Browser:**
-1. Go to [ntfy.sh](https://ntfy.sh)
-2. Click "Subscribe to topic"
-3. Enter the same topic names
-4. Allow browser notifications when prompted
-
-## 4. Add Topics to .env
-
-Add these lines to your `.env` file in the CheddahBot root:
-
-```
-NTFY_TOPIC_HUMAN_ACTION=cheddahbot-a8f3k9x2m7
-NTFY_TOPIC_ERRORS=cheddahbot-errors-p4w2j6n8
-```
-
-Replace with your actual topic names.
-
-## 5. Restart CheddahBot
-
-Kill the running instance and restart:
-
-```bash
-uv run python -m cheddahbot
-```
-
-You should see in the startup logs:
-
-```
-ntfy notifier initialized with 2 channel(s): human_action, errors
-ntfy notifier subscribed to notification bus
-```
-
-## What Gets Notified
-
-### human_action channel (high priority)
-Notifications where you need to do something:
-- Cora report finished and ready
-- Press release completed
-- Content outline ready for review
-- Content optimization completed
-- Link building pipeline finished
-- Cora report distributed to inbox
-
-### errors channel (urgent priority)
-Notifications when something went wrong:
-- ClickUp task failed or was skipped
-- AutoCora job failed
-- Link building pipeline error
-- Content pipeline error
-- Missing ClickUp field matches
-- File copy failures
-
-## Configuration
-
-Channel routing is configured in `config.yaml` under the `ntfy:` section. Each channel has:
-
-- `topic_env_var` — which env var holds the topic name
-- `categories` — notification categories to listen to (`clickup`, `autocora`, `linkbuilding`, `content`)
-- `include_patterns` — regex patterns the message must match (at least one)
-- `exclude_patterns` — regex patterns that reject the message (takes priority over include)
-- `priority` — ntfy priority level: `min`, `low`, `default`, `high`, `urgent`
-- `tags` — emoji shortcodes shown on the notification (e.g. `white_check_mark`, `rotating_light`)
-
-### Adding a New Channel
-
-1. Add a new entry under `ntfy.channels` in `config.yaml`
-2. Add the topic env var to `.env`
-3. Subscribe to the topic in your ntfy app
-4. Restart CheddahBot
-
-### Privacy
-
-The public ntfy.sh server has no authentication by default. Your topic name is the only security — use a long random string to make it unguessable. Alternatively:
-
-- Create a free ntfy.sh account and set read/write ACLs on your topics
-- Self-host ntfy (single binary) and set `server: http://localhost:8080` in config.yaml
-
-### Disabling
-
-Set `enabled: false` in the `ntfy:` section of `config.yaml`, or remove the env vars from `.env`.

docs/scheduler-refactor-notes.md

@@ -1,43 +0,0 @@
-# Scheduler Refactor Notes
-
-## Issue: AutoCora Single-Day Window (found 2026-02-27)
-
-**Symptom:** Task `86b8grf16` ("LINKS - anti vibration rubber mounts", due Feb 18) has been sitting in "to do" forever with no Cora report generated.
-
-**Root cause:** `_find_qualifying_tasks()` in `tools/autocora.py` filters tasks to **exactly one calendar day** (the `target_date`, which defaults to today). The scheduler calls this daily with `today`:
-
-```python
-today = datetime.now(UTC).strftime("%Y-%m-%d")
-result = submit_autocora_jobs(target_date=today, ctx=ctx)
-```
-
-If CheddahBot isn't running on the task's due date (or the DB is empty/wiped), the task is **permanently orphaned** — no catch-up, no retry, no visibility.
-
-**Affected task types:** All three `cora_categories` — Link Building, On Page Optimization, Content Creation.
-
-**What needs to change:** Auto-submit should also pick up overdue tasks (due date in the past, still "to do", no existing AutoCora job in KV store).
-
----
-
-## Empty Database State (found 2026-02-27)
-
-`cheddahbot.db` has zero rows in all tables (kv_store, notifications, scheduled_tasks, etc.). Either fresh DB or wiped. This means:
-- No task state tracking is happening
-- No AutoCora job submissions are recorded
-- Folder watcher has no history
-- All loops show no `last_run` timestamps
-
----
-
-## Context: Claude Scheduled Tasks
-
-Claude released scheduled tasks (2026-02-26). Need to evaluate whether parts of CheddahBot's scheduler (heartbeat, poll loop, ClickUp polling, folder watchers, AutoCora) could be replaced or augmented by Claude's native scheduling.
-
----
-
-## Additional Issues to Investigate
-
-- [ ] `auto_execute: false` on Link Building — is this intentional given the folder-watcher pipeline?
-- [ ] Folder watcher at `Z:/cora-inbox` — does this path stay accessible?
-- [ ] No dashboard/UI surfacing "tasks waiting for action" — stuck tasks are invisible
-- [ ] AutoCora loop waits 30s before first poll, then runs every 5min — but auto-submit only checks today's tasks each cycle (redundant repeated calls)

docs/task-pipeline-flows.md

@@ -50,10 +50,9 @@ CheddahBot runs 6 daemon threads. All start at boot and run until shutdown.
 | **poll** | 60 seconds | Runs cron-scheduled tasks from the database |
 | **heartbeat** | 30 minutes | Reads HEARTBEAT.md checklist, takes action if needed |
 | **clickup** | 20 minutes | Polls ClickUp for tasks to auto-execute (only Press Releases currently) |
-| **folder_watch** | 40 minutes | Scans `//PennQnap1/SHARE1/cora-inbox` for .xlsx files → triggers Link Building |
+| **folder_watch** | 60 minutes | Scans `Z:/cora-inbox` for .xlsx files → triggers Link Building |
 | **autocora** | 5 minutes | Submits Cora jobs for today's tasks + polls for results |
-| **content_watch** | 40 minutes | Scans `//PennQnap1/SHARE1/content-cora-inbox` for .xlsx files → triggers Content/OPT Phase 1 |
+| **content_watch** | 60 minutes | Scans `Z:/content-cora-inbox` for .xlsx files → triggers Content/OPT Phase 1 |
-| **cora_distribute** | 40 minutes | Scans `//PennQnap1/SHARE1/Cora-For-Human` for .xlsx files → distributes to pipeline inboxes |
 
 ---
 

pyproject.toml

@@ -16,9 +16,6 @@ dependencies = [
     "edge-tts>=6.1",
     "python-docx>=1.2.0",
     "openpyxl>=3.1.5",
-    "jinja2>=3.1.6",
-    "python-multipart>=0.0.22",
-    "sse-starlette>=3.3.3",
 ]
 
 [build-system]

scripts/clickup_feb26.py

@@ -1,94 +0,0 @@
-"""Query ClickUp 'to do' tasks tagged feb26 in OPT/LINKS/Content categories."""
-
-import sys
-from datetime import datetime, timezone
-from pathlib import Path
-
-sys.stdout.reconfigure(line_buffering=True)
-sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
-
-from cheddahbot.config import load_config
-from cheddahbot.clickup import ClickUpClient
-
-CATEGORY_PREFIXES = ("opt", "link", "content", "ai content")
-TAG_FILTER = "feb26"
-
-
-def ms_to_date(ms_str: str) -> str:
-    if not ms_str:
-        return "—"
-    try:
-        ts = int(ms_str) / 1000
-        return datetime.fromtimestamp(ts, tz=timezone.utc).strftime("%m/%d")
-    except (ValueError, OSError):
-        return "—"
-
-
-def main():
-    cfg = load_config()
-    if not cfg.clickup.api_token or not cfg.clickup.space_id:
-        print("ERROR: CLICKUP_API_TOKEN or CLICKUP_SPACE_ID not set.")
-        return
-
-    client = ClickUpClient(
-        api_token=cfg.clickup.api_token,
-        workspace_id=cfg.clickup.workspace_id,
-        task_type_field_name=cfg.clickup.task_type_field_name,
-    )
-
-    try:
-        # Fetch all 'to do' tasks across the space
-        tasks = client.get_tasks_from_space(cfg.clickup.space_id, statuses=["to do"])
-
-        # Filter by feb26 tag
-        tagged = [t for t in tasks if TAG_FILTER in [tag.lower() for tag in t.tags]]
-
-        if not tagged:
-            all_tags = set()
-            for t in tasks:
-                all_tags.update(t.tags)
-            print(f"No tasks with tag '{TAG_FILTER}'. Tags seen: {sorted(all_tags)}")
-            print(f"Total 'to do' tasks found: {len(tasks)}")
-            return
-
-        # Filter to OPT/LINKS/Content categories (by task name, Work Category, or list name)
-        def is_target_category(t):
-            name_lower = t.name.lower().strip()
-            wc = (t.custom_fields.get("Work Category") or "").lower()
-            ln = (t.list_name or "").lower()
-            for prefix in CATEGORY_PREFIXES:
-                if name_lower.startswith(prefix) or prefix in wc or prefix in ln:
-                    return True
-            return False
-
-        filtered = [t for t in tagged if is_target_category(t)]
-        skipped = [t for t in tagged if not is_target_category(t)]
-
-        # Sort by due date (oldest first), tasks with no due date go last
-        filtered.sort(key=lambda t: int(t.due_date) if t.due_date else float("inf"))
-
-        top = filtered[:10]
-
-        # Build table
-        print(f"feb26-tagged 'to do' tasks — OPT / LINKS / Content (top 10, oldest first)")
-        print(f"\n{'#':>2} | {'ID':<11} | {'Keyword/Name':<50} | {'Due':<6} | {'Customer':<25} | Tags")
-        print("-" * 120)
-        for i, t in enumerate(top, 1):
-            customer = t.custom_fields.get("Customer", "") or "—"
-            due = ms_to_date(t.due_date)
-            tags = ", ".join(t.tags)
-            name = t.name[:50]
-            print(f"{i:>2} | {t.id:<11} | {name:<50} | {due:<6} | {customer:<25} | {tags}")
-
-        print(f"\nShowing {len(top)} of {len(filtered)} OPT/LINKS/Content tasks ({len(tagged)} total feb26-tagged).")
-        if skipped:
-            print(f"\nSkipped {len(skipped)} non-OPT/LINKS/Content tasks:")
-            for t in skipped:
-                print(f"  - {t.name} ({t.id})")
-
-    finally:
-        client.close()
-
-
-if __name__ == "__main__":
-    main()

scripts/clickup_feb26_query.py

@@ -1,120 +0,0 @@
-"""Query ClickUp 'to do' tasks tagged feb26 in OPT/LINKS/Content categories."""
-
-import sys
-from pathlib import Path
-from datetime import datetime, timezone
-
-# Add project root to path
-sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
-
-from cheddahbot.config import load_config
-from cheddahbot.clickup import ClickUpClient
-
-
-def ms_to_date(ms_str: str) -> str:
-    """Convert Unix-ms timestamp string to YYYY-MM-DD."""
-    if not ms_str:
-        return "—"
-    try:
-        ts = int(ms_str) / 1000
-        return datetime.fromtimestamp(ts, tz=timezone.utc).strftime("%Y-%m-%d")
-    except (ValueError, OSError):
-        return "—"
-
-
-def main():
-    cfg = load_config()
-    if not cfg.clickup.api_token or not cfg.clickup.space_id:
-        print("ERROR: CLICKUP_API_TOKEN or CLICKUP_SPACE_ID not set.")
-        return
-
-    client = ClickUpClient(
-        api_token=cfg.clickup.api_token,
-        workspace_id=cfg.clickup.workspace_id,
-        task_type_field_name=cfg.clickup.task_type_field_name,
-    )
-
-    # Step 1: Get folders, find OPT/LINKS/Content
-    target_folders = {"opt", "links", "content"}
-    try:
-        folders = client.get_folders(cfg.clickup.space_id)
-    except Exception as e:
-        print(f"ERROR fetching folders: {e}")
-        client.close()
-        return
-
-    print(f"All folders: {[f['name'] for f in folders]}")
-
-    matched_lists = []  # (list_id, list_name, folder_name)
-    for folder in folders:
-        if folder["name"].lower() in target_folders:
-            for lst in folder["lists"]:
-                matched_lists.append((lst["id"], lst["name"], folder["name"]))
-
-    if not matched_lists:
-        print(f"No folders matching {target_folders}. Falling back to full space scan.")
-        try:
-            tasks = client.get_tasks_from_space(cfg.clickup.space_id, statuses=["to do"])
-        finally:
-            client.close()
-    else:
-        print(f"Querying lists: {[(ln, fn) for _, ln, fn in matched_lists]}")
-        tasks = []
-        for list_id, list_name, folder_name in matched_lists:
-            try:
-                batch = client.get_tasks(list_id, statuses=["to do"])
-                # Stash folder name on each task for display
-                for t in batch:
-                    t._folder = folder_name
-                tasks.extend(batch)
-            except Exception as e:
-                print(f"  Error fetching {list_name}: {e}")
-        client.close()
-
-    print(f"Total 'to do' tasks from target folders: {len(tasks)}")
-
-    # Filter by "feb26" tag (case-insensitive)
-    tagged = [t for t in tasks if any(tag.lower() == "feb26" for tag in t.tags)]
-
-    if not tagged:
-        print(f"No 'to do' tasks with 'feb26' tag found.")
-        all_tags = set()
-        for t in tasks:
-            all_tags.update(t.tags)
-        print(f"Tags found across all to-do tasks: {sorted(all_tags)}")
-        return
-
-    filtered = tagged
-
-    # Sort by due date (oldest first), tasks without due date go last
-    def sort_key(t):
-        if t.due_date:
-            return (0, int(t.due_date))
-        return (1, 0)
-
-    filtered.sort(key=sort_key)
-
-    # Take top 10
-    top10 = filtered[:10]
-
-    # Build table
-    print(f"\n## ClickUp 'to do' — feb26 tag — OPT/LINKS/Content  ({len(filtered)} total, showing top 10)\n")
-    print(f"{'#':<3} | {'ID':<12} | {'Keyword/Name':<40} | {'Due':<12} | {'Customer':<20} | Tags")
-    print(f"{'—'*3} | {'—'*12} | {'—'*40} | {'—'*12} | {'—'*20} | {'—'*15}")
-
-    for i, t in enumerate(top10, 1):
-        customer = t.custom_fields.get("Customer", "") or "—"
-        due = ms_to_date(t.due_date)
-        tags = ", ".join(t.tags) if t.tags else "—"
-        name = t.name[:38] + ".." if len(t.name) > 40 else t.name
-        print(f"{i:<3} | {t.id:<12} | {name:<40} | {due:<12} | {customer:<20} | {tags}")
-
-    print(f"\nCategory breakdown:")
-    from collections import Counter
-    cats = Counter(t.task_type for t in filtered)
-    for cat, count in cats.most_common():
-        print(f"  {cat or '(none)'}: {count}")
-
-
-if __name__ == "__main__":
-    main()

scripts/feb26_query.py

@@ -1,97 +0,0 @@
-"""Query ClickUp for feb26-tagged to-do tasks in OPT/LINKS/Content categories."""
-
-from datetime import datetime, UTC
-from cheddahbot.config import load_config
-from cheddahbot.clickup import ClickUpClient
-
-cfg = load_config()
-client = ClickUpClient(
-    api_token=cfg.clickup.api_token,
-    workspace_id=cfg.clickup.workspace_id,
-    task_type_field_name=cfg.clickup.task_type_field_name,
-)
-
-tasks = client.get_tasks_from_overall_lists(cfg.clickup.space_id, statuses=["to do"])
-client.close()
-
-# Filter: tagged feb26
-feb26 = [t for t in tasks if "feb26" in t.tags]
-
-# Filter: OPT / LINKS / Content categories (by Work Category or name prefix)
-def is_target(t):
-    cat = (t.task_type or "").lower()
-    name = t.name.upper()
-    if cat in ("on page optimization", "link building", "content creation"):
-        return True
-    if name.startswith("OPT") or name.startswith("LINKS") or name.startswith("NEW -"):
-        return True
-    return False
-
-filtered = [t for t in feb26 if is_target(t)]
-
-# Sort by due date ascending (no due date = sort last)
-def sort_key(t):
-    if t.due_date:
-        return int(t.due_date)
-    return float("inf")
-
-filtered.sort(key=sort_key)
-top10 = filtered[:10]
-
-def fmt_due(ms_str):
-    if not ms_str:
-        return "No due"
-    ts = int(ms_str) / 1000
-    return datetime.fromtimestamp(ts, tz=UTC).strftime("%b %d")
-
-def fmt_customer(t):
-    c = t.custom_fields.get("Customer", "")
-    if c and str(c) != "None":
-        return str(c)
-    return t.list_name
-
-def fmt_cat(t):
-    cat = t.task_type
-    name = t.name.upper()
-    if not cat or cat.strip() == "":
-        if name.startswith("LINKS"):
-            return "LINKS"
-        elif name.startswith("OPT"):
-            return "OPT"
-        elif name.startswith("NEW"):
-            return "Content"
-        return "?"
-    mapping = {
-        "On Page Optimization": "OPT",
-        "Link Building": "LINKS",
-        "Content Creation": "Content",
-    }
-    return mapping.get(cat, cat)
-
-def fmt_tags(t):
-    return ", ".join(t.tags) if t.tags else ""
-
-print(f"## feb26 To-Do: OPT / LINKS / Content  ({len(filtered)} total, showing top 10 oldest)")
-print()
-print("| # | ID | Keyword/Name | Due | Customer | Tags |")
-print("|---|-----|-------------|-----|----------|------|")
-for i, t in enumerate(top10, 1):
-    name = t.name[:55]
-    tid = t.id
-    due = fmt_due(t.due_date)
-    cust = fmt_customer(t)
-    tags = fmt_tags(t)
-    print(f"| {i} | {tid} | {name} | {due} | {cust} | {tags} |")
-
-if len(filtered) > 10:
-    print()
-    remaining = filtered[10:]
-    print(f"### Remaining {len(remaining)} tasks:")
-    print("| # | ID | Keyword/Name | Due | Customer | Tags |")
-    print("|---|-----|-------------|-----|----------|------|")
-    for i, t in enumerate(remaining, 11):
-        name = t.name[:55]
-        print(f"| {i} | {t.id} | {name} | {fmt_due(t.due_date)} | {fmt_customer(t)} | {fmt_tags(t)} |")
-
-print()
-print(f"*{len(filtered)} matching tasks, {len(feb26)} total feb26 tasks, {len(tasks)} total to-do*")

scripts/feb26_tasks.py

@@ -1,87 +0,0 @@
-"""Query ClickUp 'to do' tasks tagged feb26 in OPT/LINKS/Content categories."""
-
-import os
-import sys
-from datetime import datetime, timezone
-
-sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
-
-from dotenv import load_dotenv
-
-load_dotenv(os.path.join(os.path.dirname(__file__), "..", ".env"))
-
-from cheddahbot.clickup import ClickUpClient
-
-TOKEN = os.getenv("CLICKUP_API_TOKEN", "")
-SPACE_ID = os.getenv("CLICKUP_SPACE_ID", "")
-
-if not TOKEN or not SPACE_ID:
-    print("ERROR: CLICKUP_API_TOKEN and CLICKUP_SPACE_ID must be set in .env")
-    sys.exit(1)
-
-CATEGORIES = {"On Page Optimization", "Content Creation", "Link Building"}
-TAG_FILTER = "feb26"
-
-client = ClickUpClient(api_token=TOKEN, workspace_id="", task_type_field_name="Work Category")
-
-print(f"Querying ClickUp space {SPACE_ID} for 'to do' tasks...")
-tasks = client.get_tasks_from_space(SPACE_ID, statuses=["to do"])
-client.close()
-
-print(f"Total 'to do' tasks found: {len(tasks)}")
-
-# Filter by feb26 tag
-tagged = [t for t in tasks if TAG_FILTER in [tag.lower() for tag in t.tags]]
-print(f"Tasks with '{TAG_FILTER}' tag: {len(tagged)}")
-
-# Filter by Work Category (OPT / LINKS / Content)
-filtered = []
-for t in tagged:
-    cat = (t.custom_fields.get("Work Category") or t.task_type or "").strip()
-    if cat in CATEGORIES:
-        filtered.append(t)
-
-if not filtered and tagged:
-    # Show what categories exist so we can refine
-    cats_found = set()
-    for t in tagged:
-        cats_found.add(t.custom_fields.get("Work Category") or t.task_type or "(none)")
-    print(f"\nNo tasks matched categories {CATEGORIES}.")
-    print(f"Work Categories found on feb26-tagged tasks: {cats_found}")
-    print("\nShowing ALL feb26-tagged tasks instead:\n")
-    filtered = tagged
-
-# Sort by due date (oldest first), tasks without due date go last
-def sort_key(t):
-    if t.due_date:
-        return int(t.due_date)
-    return float("inf")
-
-filtered.sort(key=sort_key)
-
-# Take top 10
-top = filtered[:10]
-
-# Format table
-def fmt_due(raw_due: str) -> str:
-    if not raw_due:
-        return "—"
-    try:
-        ts = int(raw_due) / 1000
-        return datetime.fromtimestamp(ts, tz=timezone.utc).strftime("%m/%d")
-    except (ValueError, OSError):
-        return raw_due
-
-def fmt_customer(t) -> str:
-    return t.custom_fields.get("Customer", "") or "—"
-
-print(f"\n{'#':<3} | {'ID':<12} | {'Keyword/Name':<45} | {'Cat':<15} | {'Due':<6} | {'Customer':<20} | Tags")
-print("-" * 120)
-
-for i, t in enumerate(top, 1):
-    tags_str = ", ".join(t.tags)
-    name = t.name[:45]
-    cat = t.custom_fields.get("Work Category") or t.task_type or "—"
-    print(f"{i:<3} | {t.id:<12} | {name:<45} | {cat:<15} | {fmt_due(t.due_date):<6} | {fmt_customer(t):<20} | {tags_str}")
-
-print(f"\nTotal shown: {len(top)} of {len(filtered)} matching tasks")

scripts/find_pr_feb.py

@@ -1,64 +0,0 @@
-"""Find all Press Release tasks due in February 2026, any status."""
-
-import logging
-from datetime import UTC, datetime
-
-logging.basicConfig(level=logging.WARNING)
-
-from cheddahbot.config import load_config
-from cheddahbot.clickup import ClickUpClient
-import json
-
-config = load_config()
-client = ClickUpClient(
-    api_token=config.clickup.api_token,
-    workspace_id=config.clickup.workspace_id,
-    task_type_field_name=config.clickup.task_type_field_name,
-)
-
-space_id = config.clickup.space_id
-list_ids = client.get_list_ids_from_space(space_id)
-field_filter = client.discover_field_filter(
-    next(iter(list_ids)), config.clickup.task_type_field_name
-)
-
-pr_opt_id = field_filter["options"]["Press Release"]
-custom_fields_filter = json.dumps(
-    [{"field_id": field_filter["field_id"], "operator": "ANY", "value": [pr_opt_id]}]
-)
-
-# February 2026 window
-feb_start = int(datetime(2026, 2, 1, tzinfo=UTC).timestamp() * 1000)
-feb_end = int(datetime(2026, 3, 1, tzinfo=UTC).timestamp() * 1000)
-
-# Query with broad statuses, include closed
-tasks = client.get_tasks_from_space(
-    space_id,
-    custom_fields=custom_fields_filter,
-)
-
-# Filter for due in February 2026
-feb_prs = []
-for t in tasks:
-    if t.task_type != "Press Release":
-        continue
-    if not t.due_date:
-        continue
-    try:
-        due_ms = int(t.due_date)
-        if feb_start <= due_ms < feb_end:
-            feb_prs.append(t)
-    except (ValueError, TypeError):
-        continue
-
-print(f"\nPress Release tasks due in February 2026: {len(feb_prs)}\n")
-for t in feb_prs:
-    due_dt = datetime.fromtimestamp(int(t.due_date) / 1000, tz=UTC)
-    due = due_dt.strftime("%Y-%m-%d")
-    tags_str = ", ".join(t.tags) if t.tags else "(none)"
-    customer = t.custom_fields.get("Customer", "?")
-    imsurl = t.custom_fields.get("IMSURL", "")
-    print(f"  [{t.status:20s}] {t.name}")
-    print(f"    id={t.id}  due={due}  tags={tags_str}")
-    print(f"    customer={customer}  imsurl={imsurl or '(none)'}")
-    print()

scripts/find_pr_tasks.py

@@ -1,61 +0,0 @@
-"""Find all feb26-tagged Press Release tasks regardless of due date or status."""
-
-import logging
-from datetime import UTC, datetime
-
-logging.basicConfig(level=logging.INFO, format="%(asctime)s %(message)s", datefmt="%H:%M:%S")
-
-from cheddahbot.config import load_config
-from cheddahbot.clickup import ClickUpClient
-
-config = load_config()
-client = ClickUpClient(
-    api_token=config.clickup.api_token,
-    workspace_id=config.clickup.workspace_id,
-    task_type_field_name=config.clickup.task_type_field_name,
-)
-
-space_id = config.clickup.space_id
-
-# Query ALL statuses (no status filter, no due date filter) but filter by Press Release
-list_ids = client.get_list_ids_from_space(space_id)
-field_filter = client.discover_field_filter(
-    next(iter(list_ids)), config.clickup.task_type_field_name
-)
-
-import json
-pr_opt_id = field_filter["options"]["Press Release"]
-custom_fields_filter = json.dumps(
-    [{"field_id": field_filter["field_id"], "operator": "ANY", "value": [pr_opt_id]}]
-)
-
-# Get tasks with NO status filter and NO due date filter
-tasks = client.get_tasks_from_space(
-    space_id,
-    statuses=["to do", "outline approved", "in progress", "automation underway"],
-    custom_fields=custom_fields_filter,
-)
-
-# Filter for feb26 tag
-feb26_tasks = [t for t in tasks if "feb26" in t.tags]
-all_pr = [t for t in tasks if t.task_type == "Press Release"]
-
-print(f"\n{'='*70}")
-print(f"Total tasks returned: {len(tasks)}")
-print(f"Press Release tasks: {len(all_pr)}")
-print(f"feb26-tagged PR tasks: {len(feb26_tasks)}")
-print(f"{'='*70}\n")
-
-for t in all_pr:
-    due = ""
-    if t.due_date:
-        try:
-            due_dt = datetime.fromtimestamp(int(t.due_date) / 1000, tz=UTC)
-            due = due_dt.strftime("%Y-%m-%d")
-        except (ValueError, TypeError):
-            due = t.due_date
-    tags_str = ", ".join(t.tags) if t.tags else "(no tags)"
-    customer = t.custom_fields.get("Customer", "?")
-    print(f"  [{t.status:20s}] {t.name}")
-    print(f"    id={t.id}  due={due or '(none)'}  tags={tags_str}  customer={customer}")
-    print()

scripts/query_feb26.py

@@ -1,102 +0,0 @@
-"""Query ClickUp 'to do' tasks tagged 'feb26' in OPT/LINKS/Content categories."""
-
-from __future__ import annotations
-
-import os
-import sys
-from datetime import datetime, timezone
-from pathlib import Path
-
-_root = Path(__file__).resolve().parent.parent
-sys.path.insert(0, str(_root))
-
-from dotenv import load_dotenv
-load_dotenv(_root / ".env")
-
-from cheddahbot.clickup import ClickUpClient
-
-API_TOKEN = os.environ.get("CLICKUP_API_TOKEN", "")
-SPACE_ID = os.environ.get("CLICKUP_SPACE_ID", "")
-
-if not API_TOKEN:
-    sys.exit("ERROR: CLICKUP_API_TOKEN env var is required")
-if not SPACE_ID:
-    sys.exit("ERROR: CLICKUP_SPACE_ID env var is required")
-
-# Work Category values to include (case-insensitive partial match)
-CATEGORY_FILTERS = ["opt", "link", "content"]
-TAG_FILTER = "feb26"
-
-
-def ms_to_date(ms_str: str) -> str:
-    """Convert Unix-ms timestamp string to YYYY-MM-DD."""
-    if not ms_str:
-        return "—"
-    try:
-        ts = int(ms_str) / 1000
-        return datetime.fromtimestamp(ts, tz=timezone.utc).strftime("%Y-%m-%d")
-    except (ValueError, OSError):
-        return ms_str
-
-
-def main() -> None:
-    client = ClickUpClient(api_token=API_TOKEN, task_type_field_name="Work Category")
-
-    print(f"Fetching 'to do' tasks from space {SPACE_ID} ...")
-    tasks = client.get_tasks_from_overall_lists(SPACE_ID, statuses=["to do"])
-    print(f"Total 'to do' tasks: {len(tasks)}")
-
-    # Filter by feb26 tag
-    tagged = [t for t in tasks if TAG_FILTER in [tag.lower() for tag in t.tags]]
-    print(f"Tasks with '{TAG_FILTER}' tag: {len(tagged)}")
-
-    # Show all Work Category values for debugging
-    categories = set()
-    for t in tagged:
-        wc = t.custom_fields.get("Work Category", "") or ""
-        categories.add(wc)
-    print(f"Work Categories found: {categories}")
-
-    # Filter by OPT/LINKS/Content categories
-    filtered = []
-    for t in tagged:
-        wc = str(t.custom_fields.get("Work Category", "") or "").lower()
-        if any(cat in wc for cat in CATEGORY_FILTERS):
-            filtered.append(t)
-
-    print(f"After category filter (OPT/LINKS/Content): {len(filtered)}")
-
-    # Sort by due date (oldest first), tasks with no due date go last
-    def sort_key(t):
-        if t.due_date:
-            try:
-                return (0, int(t.due_date))
-            except ValueError:
-                return (1, 0)
-        return (2, 0)
-
-    filtered.sort(key=sort_key)
-
-    # Top 10
-    top10 = filtered[:10]
-
-    # Print table
-    print(f"\n{'#':>3} | {'ID':>11} | {'Keyword/Name':<45} | {'Due':>10} | {'Customer':<20} | Tags")
-    print("-" * 120)
-
-    for i, t in enumerate(top10, 1):
-        customer = t.custom_fields.get("Customer", "") or "—"
-        due = ms_to_date(t.due_date)
-        wc = t.custom_fields.get("Work Category", "") or ""
-        tags_str = ", ".join(t.tags)
-        name_display = t.name[:45] if len(t.name) > 45 else t.name
-        print(f"{i:>3} | {t.id:>11} | {name_display:<45} | {due:>10} | {customer:<20} | {tags_str}")
-
-    if not top10:
-        print("  (no matching tasks found)")
-
-    print(f"\n--- {len(filtered)} total matching tasks, showing top {len(top10)} (oldest first) ---")
-
-
-if __name__ == "__main__":
-    main()

scripts/rebuild_customer_field.py

@@ -1,149 +0,0 @@
-"""One-time script: rebuild the 'Customer' dropdown custom field in ClickUp.
-
-Steps:
-  1. Fetch all folders from the PII-Agency-SEO space
-  2. Filter out non-client folders
-  3. Create a 'Customer' dropdown field with folder names as options
-  4. For each client folder, find the 'Overall' list and set Customer on all tasks
-
-Usage:
-  DRY_RUN=1 uv run python scripts/rebuild_customer_field.py   # preview only
-  uv run python scripts/rebuild_customer_field.py              # live run
-"""
-
-from __future__ import annotations
-
-import os
-import sys
-import time
-from pathlib import Path
-
-# Allow running from repo root
-_root = Path(__file__).resolve().parent.parent
-sys.path.insert(0, str(_root))
-
-from dotenv import load_dotenv
-
-load_dotenv(_root / ".env")
-
-from cheddahbot.clickup import ClickUpClient
-
-# ── Config ──────────────────────────────────────────────────────────────────
-DRY_RUN = os.environ.get("DRY_RUN", "0") not in ("0", "false", "")
-EXCLUDED_FOLDERS = {"SEO Audits", "SEO Projects", "Business Related"}
-FIELD_NAME = "Customer"
-
-API_TOKEN = os.environ.get("CLICKUP_API_TOKEN", "")
-SPACE_ID = os.environ.get("CLICKUP_SPACE_ID", "")
-
-if not API_TOKEN:
-    sys.exit("ERROR: CLICKUP_API_TOKEN env var is required")
-if not SPACE_ID:
-    sys.exit("ERROR: CLICKUP_SPACE_ID env var is required")
-
-
-def main() -> None:
-    client = ClickUpClient(api_token=API_TOKEN)
-
-    # 1. Get folders
-    print(f"\n{'=' * 60}")
-    print(f"  Rebuild '{FIELD_NAME}' field -- Space {SPACE_ID}")
-    print(f"  Mode: {'DRY RUN' if DRY_RUN else 'LIVE'}")
-    print(f"{'=' * 60}\n")
-
-    folders = client.get_folders(SPACE_ID)
-    print(f"Found {len(folders)} folders:\n")
-
-    client_folders = []
-    for f in folders:
-        excluded = f["name"] in EXCLUDED_FOLDERS
-        marker = "  [SKIP]" if excluded else ""
-        list_names = [lst["name"] for lst in f["lists"]]
-        print(f"  {f['name']}{marker}  (lists: {', '.join(list_names) or 'none'})")
-        if not excluded:
-            client_folders.append(f)
-
-    if not client_folders:
-        sys.exit("\nNo client folders found -- nothing to do.")
-
-    option_names = sorted(f["name"] for f in client_folders)
-    print(f"\nDropdown options ({len(option_names)}): {', '.join(option_names)}")
-
-    # 2. Build a plan: folder → Overall list → tasks
-    plan: list[dict] = []  # {folder_name, list_id, tasks: [ClickUpTask]}
-    first_list_id = None
-
-    for f in client_folders:
-        overall = next((lst for lst in f["lists"] if lst["name"] == "Overall"), None)
-        if overall is None:
-            print(f"\n  WARNING: '{f['name']}' has no 'Overall' list -- skipping task update")
-            continue
-        if first_list_id is None:
-            first_list_id = overall["id"]
-        tasks = client.get_tasks(overall["id"])
-        plan.append({"folder_name": f["name"], "list_id": overall["id"], "tasks": tasks})
-
-    # 3. Print summary
-    total_tasks = sum(len(p["tasks"]) for p in plan)
-    print("\n--- Update Plan ---")
-    for p in plan:
-        print(f"  {p['folder_name']:30s}  ->  {len(p['tasks']):3d} tasks in list {p['list_id']}")
-    print(f"  {'TOTAL':30s}  ->  {total_tasks:3d} tasks")
-
-    if DRY_RUN:
-        print("\n** DRY RUN -- no changes made. Unset DRY_RUN to execute. **\n")
-        return
-
-    if first_list_id is None:
-        sys.exit("\nNo 'Overall' list found in any client folder -- cannot create field.")
-
-    # 4. Create the dropdown field
-    print(f"\nCreating '{FIELD_NAME}' dropdown on list {first_list_id} ...")
-    type_config = {
-        "options": [{"name": name, "color": None} for name in option_names],
-    }
-    client.create_custom_field(first_list_id, FIELD_NAME, "drop_down", type_config)
-    print("  Field created.")
-
-    # Brief pause for ClickUp to propagate
-    time.sleep(2)
-
-    # 5. Discover the field UUID + option IDs
-    print("Discovering field UUID and option IDs ...")
-    field_info = client.discover_field_filter(first_list_id, FIELD_NAME)
-    if field_info is None:
-        sys.exit(f"\nERROR: Could not find '{FIELD_NAME}' field after creation!")
-
-    field_id = field_info["field_id"]
-    option_map = field_info["options"]  # {name: uuid}
-    print(f"  Field ID: {field_id}")
-    print(f"  Options: {option_map}")
-
-    # 6. Set Customer field on each task
-    updated = 0
-    failed = 0
-    for p in plan:
-        folder_name = p["folder_name"]
-        opt_id = option_map.get(folder_name)
-        if not opt_id:
-            print(f"\n  WARNING: No option ID for '{folder_name}' -- skipping")
-            continue
-
-        print(f"\nUpdating {len(p['tasks'])} tasks in '{folder_name}' ...")
-        for task in p["tasks"]:
-            ok = client.set_custom_field_value(task.id, field_id, opt_id)
-            if ok:
-                updated += 1
-            else:
-                failed += 1
-                print(f"  FAILED: task {task.id} ({task.name})")
-            # Light rate-limit courtesy
-            time.sleep(0.15)
-
-    print(f"\n{'=' * 60}")
-    print(f"  Done!  Updated: {updated}  |  Failed: {failed}")
-    print(f"{'=' * 60}\n")
-
-
-if __name__ == "__main__":
-    main()

scripts/rerun_pr_tasks.py

@@ -1,144 +0,0 @@
-"""Re-run press release pipeline for specific tasks that are missing attachments."""
-
-import logging
-import sys
-import io
-
-sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding="utf-8")
-
-logging.basicConfig(
-    level=logging.INFO,
-    format="%(asctime)s [%(name)s] %(levelname)s: %(message)s",
-    datefmt="%H:%M:%S",
-    handlers=[logging.StreamHandler(stream=io.TextIOWrapper(sys.stderr.buffer, encoding="utf-8"))],
-)
-log = logging.getLogger("pr_rerun")
-
-from cheddahbot.config import load_config
-from cheddahbot.db import Database
-from cheddahbot.llm import LLMAdapter
-from cheddahbot.agent import Agent
-from cheddahbot.clickup import ClickUpClient
-
-
-TASKS_TO_RERUN = [
-    ("86b8ebfk9", "Advanced Industrial highlights medical grade plastic expertise", "Advanced Industrial"),
-]
-
-
-def bootstrap():
-    config = load_config()
-    db = Database(config.db_path)
-    llm = LLMAdapter(
-        default_model=config.chat_model,
-        openrouter_key=config.openrouter_api_key,
-        ollama_url=config.ollama_url,
-        lmstudio_url=config.lmstudio_url,
-    )
-
-    agent_cfg = config.agents[0] if config.agents else None
-    agent = Agent(config, db, llm, agent_config=agent_cfg)
-
-    try:
-        from cheddahbot.memory import MemorySystem
-        scope = agent_cfg.memory_scope if agent_cfg else ""
-        memory = MemorySystem(config, db, scope=scope)
-        agent.set_memory(memory)
-    except Exception as e:
-        log.warning("Memory not available: %s", e)
-
-    from cheddahbot.tools import ToolRegistry
-    tools = ToolRegistry(config, db, agent)
-    agent.set_tools(tools)
-
-    try:
-        from cheddahbot.skills import SkillRegistry
-        skills = SkillRegistry(config.skills_dir)
-        agent.set_skills_registry(skills)
-    except Exception as e:
-        log.warning("Skills not available: %s", e)
-
-    return config, db, agent, tools
-
-
-def run_task(agent, tools, config, client, task_id, task_name, customer):
-    """Execute write_press_releases for a specific task."""
-    # Build args matching the field_mapping from config
-    args = {
-        "topic": task_name,
-        "company_name": customer,
-        "clickup_task_id": task_id,
-    }
-
-    # Also fetch IMSURL from the task
-    import httpx as _httpx
-    resp = _httpx.get(
-        f"https://api.clickup.com/api/v2/task/{task_id}",
-        headers={"Authorization": config.clickup.api_token},
-        timeout=30.0,
-    )
-    task_data = resp.json()
-    for cf in task_data.get("custom_fields", []):
-        if cf["name"] == "IMSURL":
-            val = cf.get("value")
-            if val:
-                args["url"] = val
-        elif cf["name"] == "SocialURL":
-            val = cf.get("value")
-            if val:
-                args["branded_url"] = val
-
-    log.info("=" * 70)
-    log.info("EXECUTING: %s", task_name)
-    log.info("  Task ID:      %s", task_id)
-    log.info("  Customer:     %s", customer)
-    log.info("  Args:         %s", {k: v for k, v in args.items() if k != "clickup_task_id"})
-    log.info("=" * 70)
-
-    try:
-        result = tools.execute("write_press_releases", args)
-
-        if result.startswith("Skipped:") or result.startswith("Error:"):
-            log.error("Task skipped/errored: %s", result[:500])
-            return False
-
-        log.info("Task completed!")
-        # Print first 500 chars of result
-        print(f"\n--- Result for {task_name} ---")
-        print(result[:1000])
-        print("--- End ---\n")
-        return True
-
-    except Exception as e:
-        log.error("Task failed: %s", e, exc_info=True)
-        return False
-
-
-def main():
-    log.info("Bootstrapping CheddahBot...")
-    config, db, agent, tools = bootstrap()
-
-    client = ClickUpClient(
-        api_token=config.clickup.api_token,
-        workspace_id=config.clickup.workspace_id,
-        task_type_field_name=config.clickup.task_type_field_name,
-    )
-
-    log.info("Will re-run %d tasks", len(TASKS_TO_RERUN))
-
-    results = []
-    for i, (task_id, name, customer) in enumerate(TASKS_TO_RERUN):
-        log.info("\n>>> Task %d/%d <<<", i + 1, len(TASKS_TO_RERUN))
-        success = run_task(agent, tools, config, client, task_id, name, customer)
-        results.append((name, success))
-
-    print(f"\n{'=' * 70}")
-    print("RESULTS SUMMARY")
-    print(f"{'=' * 70}")
-    for name, success in results:
-        status = "OK" if success else "FAILED"
-        print(f"  [{status}] {name}")
-
-
-if __name__ == "__main__":
-    main()

scripts/run_pr_pipeline.py

@@ -1,241 +0,0 @@
-"""Run the press-release pipeline for up to N ClickUp tasks.
-
-Usage:
-    uv run python scripts/run_pr_pipeline.py          # discover + execute up to 3
-    uv run python scripts/run_pr_pipeline.py --dry-run # discover only, don't execute
-    uv run python scripts/run_pr_pipeline.py --max 1   # execute only 1 task
-"""
-
-import argparse
-import logging
-import sys
-from datetime import UTC, datetime
-
-logging.basicConfig(
-    level=logging.INFO,
-    format="%(asctime)s [%(name)s] %(levelname)s: %(message)s",
-    datefmt="%H:%M:%S",
-)
-log = logging.getLogger("pr_pipeline")
-
-# ── Bootstrap CheddahBot (config, db, agent, tools) ─────────────────────
-
-from cheddahbot.config import load_config
-from cheddahbot.db import Database
-from cheddahbot.llm import LLMAdapter
-from cheddahbot.agent import Agent
-from cheddahbot.clickup import ClickUpClient
-
-
-def bootstrap():
-    """Set up config, db, agent, and tool registry — same as __main__.py."""
-    config = load_config()
-    db = Database(config.db_path)
-    llm = LLMAdapter(
-        default_model=config.chat_model,
-        openrouter_key=config.openrouter_api_key,
-        ollama_url=config.ollama_url,
-        lmstudio_url=config.lmstudio_url,
-    )
-
-    agent_cfg = config.agents[0] if config.agents else None
-    agent = Agent(config, db, llm, agent_config=agent_cfg)
-
-    # Memory
-    try:
-        from cheddahbot.memory import MemorySystem
-        scope = agent_cfg.memory_scope if agent_cfg else ""
-        memory = MemorySystem(config, db, scope=scope)
-        agent.set_memory(memory)
-    except Exception as e:
-        log.warning("Memory not available: %s", e)
-
-    # Tools
-    from cheddahbot.tools import ToolRegistry
-    tools = ToolRegistry(config, db, agent)
-    agent.set_tools(tools)
-
-    # Skills
-    try:
-        from cheddahbot.skills import SkillRegistry
-        skills = SkillRegistry(config.skills_dir)
-        agent.set_skills_registry(skills)
-    except Exception as e:
-        log.warning("Skills not available: %s", e)
-
-    return config, db, agent, tools
-
-
-def discover_pr_tasks(config):
-    """Poll ClickUp for Press Release tasks — same logic as scheduler._poll_clickup()."""
-    client = ClickUpClient(
-        api_token=config.clickup.api_token,
-        workspace_id=config.clickup.workspace_id,
-        task_type_field_name=config.clickup.task_type_field_name,
-    )
-    space_id = config.clickup.space_id
-    skill_map = config.clickup.skill_map
-
-    if not space_id:
-        log.error("No space_id configured")
-        return [], client
-
-    # Discover field filter (Work Category UUID + options)
-    list_ids = client.get_list_ids_from_space(space_id)
-    if not list_ids:
-        log.error("No lists found in space %s", space_id)
-        return [], client
-
-    first_list = next(iter(list_ids))
-    field_filter = client.discover_field_filter(
-        first_list, config.clickup.task_type_field_name
-    )
-
-    # Build custom fields filter for API query
-    custom_fields_filter = None
-    if field_filter and field_filter.get("options"):
-        import json
-        field_id = field_filter["field_id"]
-        options = field_filter["options"]
-        # Only Press Release
-        pr_opt_id = options.get("Press Release")
-        if pr_opt_id:
-            custom_fields_filter = json.dumps(
-                [{"field_id": field_id, "operator": "ANY", "value": [pr_opt_id]}]
-            )
-            log.info("Filtering for Press Release option ID: %s", pr_opt_id)
-        else:
-            log.warning("'Press Release' not found in Work Category options: %s", list(options.keys()))
-            return [], client
-
-    # Due date window (3 weeks)
-    now_ms = int(datetime.now(UTC).timestamp() * 1000)
-    due_date_lt = now_ms + (3 * 7 * 24 * 60 * 60 * 1000)
-
-    tasks = client.get_tasks_from_space(
-        space_id,
-        statuses=config.clickup.poll_statuses,
-        due_date_lt=due_date_lt,
-        custom_fields=custom_fields_filter,
-    )
-
-    # Client-side filter: must be Press Release + have due date in window
-    pr_tasks = []
-    for task in tasks:
-        if task.task_type != "Press Release":
-            continue
-        if not task.due_date:
-            continue
-        try:
-            if int(task.due_date) > due_date_lt:
-                continue
-        except (ValueError, TypeError):
-            continue
-        pr_tasks.append(task)
-
-    return pr_tasks, client
-
-
-def execute_task(agent, tools, config, client, task):
-    """Execute a single PR task — same logic as scheduler._execute_task()."""
-    skill_map = config.clickup.skill_map
-    mapping = skill_map.get("Press Release", {})
-    tool_name = mapping.get("tool", "write_press_releases")
-
-    task_id = task.id
-
-    # Build tool args from field mapping
-    field_mapping = mapping.get("field_mapping", {})
-    args = {}
-    for tool_param, source in field_mapping.items():
-        if source == "task_name":
-            args[tool_param] = task.name
-        elif source == "task_description":
-            args[tool_param] = task.custom_fields.get("description", "")
-        else:
-            args[tool_param] = task.custom_fields.get(source, "")
-
-    args["clickup_task_id"] = task_id
-
-    log.info("=" * 70)
-    log.info("EXECUTING: %s", task.name)
-    log.info("  Task ID:      %s", task_id)
-    log.info("  Tool:         %s", tool_name)
-    log.info("  Args:         %s", {k: v for k, v in args.items() if k != "clickup_task_id"})
-    log.info("=" * 70)
-
-    # Move to "automation underway"
-    client.update_task_status(task_id, config.clickup.automation_status)
-
-    try:
-        result = tools.execute(tool_name, args)
-
-        if result.startswith("Skipped:") or result.startswith("Error:"):
-            log.error("Task skipped/errored: %s", result[:500])
-            client.add_comment(
-                task_id,
-                f"⚠️ CheddahBot could not execute this task.\n\n{result[:2000]}",
-            )
-            client.update_task_status(task_id, config.clickup.error_status)
-            return False
-
-        log.info("Task completed successfully!")
-        log.info("Result preview:\n%s", result[:1000])
-        return True
-
-    except Exception as e:
-        log.error("Task failed with exception: %s", e, exc_info=True)
-        client.add_comment(
-            task_id,
-            f"❌ CheddahBot failed to complete this task.\n\nError: {str(e)[:2000]}",
-        )
-        client.update_task_status(task_id, config.clickup.error_status)
-        return False
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Run PR pipeline from ClickUp")
-    parser.add_argument("--dry-run", action="store_true", help="Discover only, don't execute")
-    parser.add_argument("--max", type=int, default=3, help="Max tasks to execute (default: 3)")
-    args = parser.parse_args()
-
-    log.info("Bootstrapping CheddahBot...")
-    config, db, agent, tools = bootstrap()
-
-    log.info("Polling ClickUp for Press Release tasks...")
-    pr_tasks, client = discover_pr_tasks(config)
-
-    if not pr_tasks:
-        log.info("No Press Release tasks found in statuses %s", config.clickup.poll_statuses)
-        return
-
-    log.info("Found %d Press Release task(s):", len(pr_tasks))
-    for i, task in enumerate(pr_tasks):
-        status_str = f"status={task.status}" if hasattr(task, "status") else ""
-        log.info("  %d. %s (id=%s) %s", i + 1, task.name, task.id, status_str)
-        log.info("     Custom fields: %s", task.custom_fields)
-
-    if args.dry_run:
-        log.info("Dry run — not executing. Use without --dry-run to execute.")
-        return
-
-    # Execute up to --max tasks
-    to_run = pr_tasks[: args.max]
-    log.info("Will execute %d task(s) (max=%d)", len(to_run), args.max)
-
-    results = []
-    for i, task in enumerate(to_run):
-        log.info("\n>>> Task %d/%d <<<", i + 1, len(to_run))
-        success = execute_task(agent, tools, config, client, task)
-        results.append((task.name, success))
-
-    log.info("\n" + "=" * 70)
-    log.info("RESULTS SUMMARY")
-    log.info("=" * 70)
-    for name, success in results:
-        status = "OK" if success else "FAILED"
-        log.info("  [%s] %s", status, name)
-
-
-if __name__ == "__main__":
-    main()

start.sh

@@ -1,3 +0,0 @@
-#!/usr/bin/env bash
-cd "$(dirname "$0")"
-exec uv run python -m cheddahbot

tests/test_clickup_tools.py

@@ -10,7 +10,6 @@ from cheddahbot.tools.clickup_tool import (
     clickup_query_tasks,
     clickup_reset_task,
     clickup_task_status,
-    get_active_tasks,
 )
 
 
@@ -143,38 +142,3 @@ class TestClickupResetTask:
 
         result = clickup_reset_task(task_id="t1", ctx=_make_ctx())
         assert "Error" in result
-
-
-class TestGetActiveTasks:
-    def test_no_scheduler(self):
-        result = get_active_tasks(ctx={"config": MagicMock()})
-        assert "not available" in result.lower()
-
-    def test_nothing_running(self):
-        scheduler = MagicMock()
-        scheduler.get_active_executions.return_value = {}
-        scheduler.get_loop_timestamps.return_value = {"clickup": None, "folder_watch": None}
-
-        result = get_active_tasks(ctx={"scheduler": scheduler})
-        assert "No tasks actively executing" in result
-        assert "Safe to restart: Yes" in result
-
-    def test_tasks_running(self):
-        from datetime import UTC, datetime, timedelta
-
-        scheduler = MagicMock()
-        scheduler.get_active_executions.return_value = {
-            "t1": {
-                "name": "Press Release for Acme",
-                "tool": "write_press_releases",
-                "started_at": datetime.now(UTC) - timedelta(minutes=5),
-                "thread": "clickup_thread",
-            }
-        }
-        scheduler.get_loop_timestamps.return_value = {"clickup": datetime.now(UTC).isoformat()}
-
-        result = get_active_tasks(ctx={"scheduler": scheduler})
-        assert "Active Executions (1)" in result
-        assert "Press Release for Acme" in result
-        assert "write_press_releases" in result
-        assert "Safe to restart: No" in result

tests/test_cora_distribute.py

@@ -16,7 +16,6 @@ class FakeTask:
     id: str = "fake_id"
     name: str = ""
     task_type: str = ""
-    status: str = "running cora"
     custom_fields: dict = field(default_factory=dict)
 
 

tests/test_linkbuilding.py

@@ -227,36 +227,23 @@ class TestFuzzyKeywordMatch:
     def test_exact_match(self):
         assert _fuzzy_keyword_match("precision cnc", "precision cnc") is True
 
-    def test_no_match_without_llm(self):
+    def test_substring_match_a_in_b(self):
-        """Without an llm_check, non-exact strings return False."""
+        assert _fuzzy_keyword_match("cnc machining", "precision cnc machining services") is True
-        assert _fuzzy_keyword_match("shaft", "shafts") is False
+
-        assert _fuzzy_keyword_match("shaft manufacturing", "custom shaft manufacturing") is False
+    def test_substring_match_b_in_a(self):
+        assert _fuzzy_keyword_match("precision cnc machining services", "cnc machining") is True
+
+    def test_word_overlap(self):
+        assert _fuzzy_keyword_match("precision cnc machining", "cnc machining precision") is True
+
+    def test_no_match(self):
+        assert _fuzzy_keyword_match("precision cnc", "web design agency") is False
 
     def test_empty_strings(self):
         assert _fuzzy_keyword_match("", "test") is False
         assert _fuzzy_keyword_match("test", "") is False
         assert _fuzzy_keyword_match("", "") is False
 
-    def test_llm_check_called_on_mismatch(self):
-        """When strings differ, llm_check is called and its result is returned."""
-        llm_yes = lambda a, b: True
-        llm_no = lambda a, b: False
-
-        assert _fuzzy_keyword_match("shaft", "shafts", llm_check=llm_yes) is True
-        assert _fuzzy_keyword_match("shaft", "shafts", llm_check=llm_no) is False
-
-    def test_llm_check_not_called_on_exact(self):
-        """Exact match should not call llm_check."""
-        def boom(a, b):
-            raise AssertionError("should not be called")
-
-        assert _fuzzy_keyword_match("shaft", "shaft", llm_check=boom) is True
-
-    def test_no_substring_match_without_llm(self):
-        """Substring matching is gone — different keywords must not match."""
-        assert _fuzzy_keyword_match("shaft manufacturing", "custom shaft manufacturing") is False
-        assert _fuzzy_keyword_match("cnc machining", "precision cnc machining services") is False
-
 
 class TestNormalizeForMatch:
     def test_lowercase_and_strip(self):

tests/test_scheduler.py

@@ -232,62 +232,3 @@ class TestFieldFilterDiscovery:
         mock_client.discover_field_filter.reset_mock()
         scheduler._poll_clickup()
         mock_client.discover_field_filter.assert_not_called()
-
-
-class TestActiveExecutions:
-    """Test the active execution registry."""
-
-    def test_register_and_get(self, tmp_db):
-        config = _FakeConfig()
-        scheduler = Scheduler(config, tmp_db, MagicMock())
-
-        scheduler._register_execution("t1", "Task One", "write_press_releases")
-        active = scheduler.get_active_executions()
-
-        assert "t1" in active
-        assert active["t1"]["name"] == "Task One"
-        assert active["t1"]["tool"] == "write_press_releases"
-        assert "started_at" in active["t1"]
-        assert "thread" in active["t1"]
-
-    def test_unregister(self, tmp_db):
-        config = _FakeConfig()
-        scheduler = Scheduler(config, tmp_db, MagicMock())
-
-        scheduler._register_execution("t1", "Task One", "write_press_releases")
-        scheduler._unregister_execution("t1")
-        assert scheduler.get_active_executions() == {}
-
-    def test_unregister_nonexistent_is_noop(self, tmp_db):
-        config = _FakeConfig()
-        scheduler = Scheduler(config, tmp_db, MagicMock())
-
-        # Should not raise
-        scheduler._unregister_execution("nonexistent")
-        assert scheduler.get_active_executions() == {}
-
-    def test_multiple_executions(self, tmp_db):
-        config = _FakeConfig()
-        scheduler = Scheduler(config, tmp_db, MagicMock())
-
-        scheduler._register_execution("t1", "Task One", "write_press_releases")
-        scheduler._register_execution("t2", "Task Two", "run_cora_backlinks")
-        active = scheduler.get_active_executions()
-
-        assert len(active) == 2
-        assert "t1" in active
-        assert "t2" in active
-
-    def test_get_returns_snapshot(self, tmp_db):
-        """get_active_executions returns a copy, not a reference."""
-        config = _FakeConfig()
-        scheduler = Scheduler(config, tmp_db, MagicMock())
-
-        scheduler._register_execution("t1", "Task One", "tool_a")
-        snapshot = scheduler.get_active_executions()
-        scheduler._unregister_execution("t1")
-
-        # Snapshot should still have t1
-        assert "t1" in snapshot
-        # But live state should be empty
-        assert scheduler.get_active_executions() == {}

uv.lock

@@ -325,16 +325,13 @@ dependencies = [
     { name = "edge-tts" },
     { name = "gradio" },
     { name = "httpx" },
-    { name = "jinja2" },
     { name = "numpy" },
     { name = "openai" },
     { name = "openpyxl" },
     { name = "python-docx" },
     { name = "python-dotenv" },
-    { name = "python-multipart" },
     { name = "pyyaml" },
     { name = "sentence-transformers" },
-    { name = "sse-starlette" },
 ]
 
 [package.dev-dependencies]
@@ -360,16 +357,13 @@ requires-dist = [
     { name = "edge-tts", specifier = ">=6.1" },
     { name = "gradio", specifier = ">=5.0" },
     { name = "httpx", specifier = ">=0.27" },
-    { name = "jinja2", specifier = ">=3.1.6" },
     { name = "numpy", specifier = ">=1.24" },
     { name = "openai", specifier = ">=1.30" },
     { name = "openpyxl", specifier = ">=3.1.5" },
     { name = "python-docx", specifier = ">=1.2.0" },
     { name = "python-dotenv", specifier = ">=1.0" },
-    { name = "python-multipart", specifier = ">=0.0.22" },
     { name = "pyyaml", specifier = ">=6.0" },
     { name = "sentence-transformers", specifier = ">=3.0" },
-    { name = "sse-starlette", specifier = ">=3.3.3" },
 ]
 
 [package.metadata.requires-dev]
@@ -2562,19 +2556,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/46/2c/1462b1d0a634697ae9e55b3cecdcb64788e8b7d63f54d923fcd0bb140aed/soupsieve-2.8.3-py3-none-any.whl", hash = "sha256:ed64f2ba4eebeab06cc4962affce381647455978ffc1e36bb79a545b91f45a95", size = 37016, upload-time = "2026-01-20T04:27:01.012Z" },
 ]
 
-[[package]]
-name = "sse-starlette"
-version = "3.3.3"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "anyio" },
-    { name = "starlette" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/14/2f/9223c24f568bb7a0c03d751e609844dce0968f13b39a3f73fbb3a96cd27a/sse_starlette-3.3.3.tar.gz", hash = "sha256:72a95d7575fd5129bd0ae15275ac6432bb35ac542fdebb82889c24bb9f3f4049", size = 32420, upload-time = "2026-03-17T20:05:55.529Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/78/e2/b8cff57a67dddf9a464d7e943218e031617fb3ddc133aeeb0602ff5f6c85/sse_starlette-3.3.3-py3-none-any.whl", hash = "sha256:c5abb5082a1cc1c6294d89c5290c46b5f67808cfdb612b7ec27e8ba061c22e8d", size = 14329, upload-time = "2026-03-17T20:05:54.35Z" },
-]
-
 [[package]]
 name = "starlette"
 version = "0.52.1"
@@ -2741,12 +2722,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/0f/8b/4b61d6e13f7108f36910df9ab4b58fd389cc2520d54d81b88660804aad99/torch-2.10.0-2-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:418997cb02d0a0f1497cf6a09f63166f9f5df9f3e16c8a716ab76a72127c714f", size = 79423467, upload-time = "2026-02-10T21:44:48.711Z" },
     { url = "https://files.pythonhosted.org/packages/d3/54/a2ba279afcca44bbd320d4e73675b282fcee3d81400ea1b53934efca6462/torch-2.10.0-2-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:13ec4add8c3faaed8d13e0574f5cd4a323c11655546f91fbe6afa77b57423574", size = 79498202, upload-time = "2026-02-10T21:44:52.603Z" },
     { url = "https://files.pythonhosted.org/packages/ec/23/2c9fe0c9c27f7f6cb865abcea8a4568f29f00acaeadfc6a37f6801f84cb4/torch-2.10.0-2-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:e521c9f030a3774ed770a9c011751fb47c4d12029a3d6522116e48431f2ff89e", size = 79498254, upload-time = "2026-02-10T21:44:44.095Z" },
-    { url = "https://files.pythonhosted.org/packages/36/ab/7b562f1808d3f65414cd80a4f7d4bb00979d9355616c034c171249e1a303/torch-2.10.0-3-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:ac5bdcbb074384c66fa160c15b1ead77839e3fe7ed117d667249afce0acabfac", size = 915518691, upload-time = "2026-03-11T14:15:43.147Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/7a/abada41517ce0011775f0f4eacc79659bc9bc6c361e6bfe6f7052a6b9363/torch-2.10.0-3-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:98c01b8bb5e3240426dcde1446eed6f40c778091c8544767ef1168fc663a05a6", size = 915622781, upload-time = "2026-03-11T14:17:11.354Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/c6/4dfe238342ffdcec5aef1c96c457548762d33c40b45a1ab7033bb26d2ff2/torch-2.10.0-3-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:80b1b5bfe38eb0e9f5ff09f206dcac0a87aadd084230d4a36eea5ec5232c115b", size = 915627275, upload-time = "2026-03-11T14:16:11.325Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/f0/72bf18847f58f877a6a8acf60614b14935e2f156d942483af1ffc081aea0/torch-2.10.0-3-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:46b3574d93a2a8134b3f5475cfb98e2eb46771794c57015f6ad1fb795ec25e49", size = 915523474, upload-time = "2026-03-11T14:17:44.422Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/39/590742415c3030551944edc2ddc273ea1fdfe8ffb2780992e824f1ebee98/torch-2.10.0-3-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:b1d5e2aba4eb7f8e87fbe04f86442887f9167a35f092afe4c237dfcaaef6e328", size = 915632474, upload-time = "2026-03-11T14:15:13.666Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/8e/34949484f764dde5b222b7fe3fede43e4a6f0da9d7f8c370bb617d629ee2/torch-2.10.0-3-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:0228d20b06701c05a8f978357f657817a4a63984b0c90745def81c18aedfa591", size = 915523882, upload-time = "2026-03-11T14:14:46.311Z" },
     { url = "https://files.pythonhosted.org/packages/78/89/f5554b13ebd71e05c0b002f95148033e730d3f7067f67423026cc9c69410/torch-2.10.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:3282d9febd1e4e476630a099692b44fdc214ee9bf8ee5377732d9d9dfe5712e4", size = 145992610, upload-time = "2026-01-21T16:25:26.327Z" },
     { url = "https://files.pythonhosted.org/packages/ae/30/a3a2120621bf9c17779b169fc17e3dc29b230c29d0f8222f499f5e159aa8/torch-2.10.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:a2f9edd8dbc99f62bc4dfb78af7bf89499bca3d753423ac1b4e06592e467b763", size = 915607863, upload-time = "2026-01-21T16:25:06.696Z" },
     { url = "https://files.pythonhosted.org/packages/6f/3d/c87b33c5f260a2a8ad68da7147e105f05868c281c63d65ed85aa4da98c66/torch-2.10.0-cp311-cp311-win_amd64.whl", hash = "sha256:29b7009dba4b7a1c960260fc8ac85022c784250af43af9fb0ebafc9883782ebd", size = 113723116, upload-time = "2026-01-21T16:25:21.916Z" },