bulk-analyse

name	bulk-analyse
description	Process every unanalysed .eml file in the user's corpus inputs/emails/ directory through the full analyse-email pipeline. Skips files that already have a corresponding analysis report. Use when the user has dropped many .eml files into the corpus at once.

bulk-analyse

Drain queue. Runs analyse-email on every .eml in the corpus that doesn't already have a report.

Resolve corpus path

Standard resolver.

Steps

1. List <corpus>/inputs/emails/*.eml (skip the sanitised/ subdirectory).

2. For each file, check <corpus>/outputs/analyses/<basename>-analysis.md. If it exists, skip.

3. For each unanalysed file, run the same pipeline as analyse-email/SKILL.md — don't re-implement, refer to that skill.

4. Print a summary table:

   • Number processed, number skipped (already analysed)
   • Total tracking pixels detected across the batch
   • Net new senders added to the blocklist
   • Top 5 indicator types observed

Performance

If the batch is > 30 files, prompt the user once: "This will run the analyst agent N times. Continue?". Otherwise just run.