Use single Cora-For-Human folder for all xlsx reads

All handlers (BLM, content, OPT) now read Cora xlsx from one shared
folder (//PennQnap1/SHARE1/Cora-For-Human). Files stay in place after
processing so multiple task types can share the same report.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

clickup_runner/README.md

@@ -91,9 +91,13 @@ clickup:
 autocora:
   jobs_dir: "//PennQnap1/SHARE1/AutoCora/jobs"
   results_dir: "//PennQnap1/SHARE1/AutoCora/results"
-  xlsx_dir: "//PennQnap1/SHARE1/Cora72-for-macro"
+  xlsx_dir: "//PennQnap1/SHARE1/Cora-For-Human"
   poll_interval_seconds: 120
 
+blm:
+  blm_dir: "E:/dev/Big-Link-Man"
+  timeout_seconds: 1800
+
 nas:
   generated_dir: "//PennQnap1/SHARE1/generated"
 
@@ -241,6 +245,28 @@ At the start of each cycle, the runner scans the results directory:
 If a task at `run_cora` stage already has an `.xlsx` attachment, the runner
 skips Cora submission and advances directly to the next stage.
 
+## BLM Handler (Link Building build stage)
+
+The Link Building `build` stage runs Big-Link-Man directly -- no Claude needed.
+
+1. Sets status to "AI Working"
+2. Looks for a Cora `.xlsx` in `//PennQnap1/SHARE1/Cora-For-Human/` matching the task keyword
+3. Runs `ingest-cora` via BLM's own venv Python (`E:/dev/Big-Link-Man/.venv/Scripts/python.exe`)
+4. Runs `generate-batch` with `--continue-on-error`
+5. On success: advances Stage to `final`, posts summary comment, unchecks Delegate to Claude
+6. On failure: sets Error checkbox, posts structured error comment
+
+BLM credentials are injected from `BLM_USERNAME` and `BLM_PASSWORD` env vars.
+
+## Cora .xlsx Files
+
+All Cora reports live in **one folder**: `//PennQnap1/SHARE1/Cora-For-Human/`
+
+- AutoCora drops finished reports here
+- All handlers (BLM, content, OPT) read from this folder by keyword match
+- Files are **not moved** after processing -- they stay in place so multiple task types can use the same report
+- A separate cleanup script can move old files to `processed/` when no open tasks match the keyword
+
 ## Logs
 
 - Console output: INFO level

clickup_runner/__main__.py

@@ -464,14 +464,14 @@ def _dispatch_blm(
     client.update_task_status(task.id, cfg.clickup.ai_working_status)
 
     # 2. Find the Cora xlsx
-    xlsx_path = find_cora_xlsx(keyword, cfg.blm.cora_inbox)
+    xlsx_path = find_cora_xlsx(keyword, cfg.autocora.xlsx_dir)
     if not xlsx_path:
         _handle_dispatch_error(
             client, cfg, db, task, run_id,
             error="No Cora xlsx found for keyword '%s' in %s"
-            % (keyword, cfg.blm.cora_inbox),
+            % (keyword, cfg.autocora.xlsx_dir),
             fix="Check that the Cora report exists in %s, then re-check Delegate to Claude."
-            % cfg.blm.cora_inbox,
+            % cfg.autocora.xlsx_dir,
         )
         return
 

clickup_runner/config.py

@@ -36,14 +36,13 @@ class ClickUpConfig:
 class AutoCoraConfig:
     jobs_dir: str = "//PennQnap1/SHARE1/AutoCora/jobs"
     results_dir: str = "//PennQnap1/SHARE1/AutoCora/results"
-    xlsx_dir: str = "//PennQnap1/SHARE1/Cora72-for-macro"
+    xlsx_dir: str = "//PennQnap1/SHARE1/Cora-For-Human"
     poll_interval_seconds: int = 120
 
 
 @dataclass
 class BLMConfig:
     blm_dir: str = "E:/dev/Big-Link-Man"
-    cora_inbox: str = "//PennQnap1/SHARE1/cora-inbox"
     timeout_seconds: int = 1800  # 30 minutes
 
 

skills/runner_linkbuilding.md

@@ -1,69 +0,0 @@
-# Link Building -- Build Stage
-
-Run the Big-Link-Man (BLM) CLI to ingest a Cora report and generate tiered backlink content.
-
-## What You Have
-
-- A Cora `.xlsx` report on the NAS at `//PennQnap1/SHARE1/Cora72-for-macro/`
-- The task's keyword, target URL, and optional CLI flags in the Task Context below
-
-## Steps
-
-### 1. Find the Cora .xlsx File
-
-Search `//PennQnap1/SHARE1/Cora72-for-macro/` for a file matching the task keyword. The filename is a slugified version of the keyword (e.g., "gearbox oil sight glass" -> `gearbox_oil_sight_glass.xlsx`). Use glob/ls to find it. If multiple matches exist, pick the most recent.
-
-### 2. Run ingest-cora
-
-Run the BLM CLI using its own venv Python (NOT the system Python or CheddahBot's venv):
-
-```bash
-E:/dev/Big-Link-Man/.venv/Scripts/python.exe main.py ingest-cora \
-  -f "<path_to_xlsx>" \
-  -n "<keyword>" \
-  -m "<target_url>" \
-  <additional_cli_flags>
-```
-
-- Working directory: `E:/dev/Big-Link-Man`
-- Always pass `-m` with the target URL (prevents interactive prompts)
-- If BrandedPlusRatio is provided and not 0.7, add `-bp <value>`
-- If CustomAnchors is provided, add `-a "<anchors>"`
-- If CLIFlags is provided, append them as-is (e.g. `--tier1-count 6`)
-- BLM credentials are in env vars `BLM_USERNAME` and `BLM_PASSWORD` -- pass them as `-u` and `-p`
-
-**Expected output:** Look for lines like:
-- `Success: Project 'name' created (ID: 42)`
-- `Job file created: jobs/some-file.json`
-
-If exit code is non-zero, stop and report the error including stdout and stderr.
-
-### 3. Run generate-batch
-
-Using the job file path from step 2:
-
-```bash
-E:/dev/Big-Link-Man/.venv/Scripts/python.exe main.py generate-batch \
-  -j "<job_file_path>" \
-  --continue-on-error
-```
-
-- Working directory: `E:/dev/Big-Link-Man`
-- Always include `--continue-on-error`
-- Pass `-u` and `-p` credentials from env vars
-- This step can take several minutes
-
-If exit code is non-zero, stop and report the error.
-
-### 4. Collect Output
-
-After generate-batch completes, the generated content files are in Big-Link-Man's output directory. Copy the relevant output files to the current working directory so they get uploaded to ClickUp.
-
-Look in `E:/dev/Big-Link-Man/output/` for the project folder matching the keyword. Copy all `.md`, `.txt`, and `.html` files from there to the current working directory.
-
-## Important
-
-- NEVER use `uv run` or CheddahBot's Python -- always use BLM's venv at `E:/dev/Big-Link-Man/.venv/Scripts/python.exe`
-- If the venv doesn't exist, stop and report the error
-- Do not modify any BLM source code
-- Do not create subdirectories in the working directory