broad-sweep

name	broad-sweep
description	Phase 1 — Run the chunks x issue_spotters cross-product to find candidate issues. Uses edsl for parallel LLM review.
user-invocable	true
allowed-tools	Bash, Read, Write, Glob, Grep

Phase 1: Broad Sweep

Run from the repo root. Requires .review/config.json, .review/parsed/, .review/symbol_table.json, and .review/issue_spotters/ from Phase 0.

How it works

The broad sweep runs a cross-product of chunks x issue_spotters:

• Chunks: The parsed manuscript is split into overlapping ~800-word windows
• Issue spotters: Named markdown files in .review/issue_spotters/ describing what to look for

Each (chunk, spotter) pair is one scenario in the edsl survey. The result is either:

• issue.json — the spotter found something in that chunk
• no_issue.json — the spotter checked and found nothing

Execute the broad sweep

Look for the roboree scripts path in the system prompt (it will say "roboree scripts are at: "). Use that path:

python <r2-scripts-path>/broad_sweep.py \
  --parsed-dir .review/parsed/ \
  --symbol-table .review/symbol_table.json \
  --spotters-dir .review/issue_spotters/ \
  --output-dir .review/chunks/ \
  --config .review/config.json