challenge

Run the challenger agent against your current changes to simulate the lead reviewer's feedback before submitting. Requires a forge analysis with reviewer patterns.

Simulate the lead reviewer's feedback on the current changes. Uses reviewer patterns from the forge analysis.

Before Starting

  1. Read .context/forge-analysis.md for reviewer patterns
  2. If it doesn't exist, tell the user to run /claude-army:analyze first

Process

  1. Run git status and git diff --name-only to find all changes
  2. Read every changed/new file
  3. Apply reviewer checks in tier order (Tier 1 first, most likely to be raised)
  4. For factual claims (thresholds, tax classification, etc.): WebSearch to verify
  5. For source URLs: WebFetch to confirm accessibility

Output

Present findings as the lead reviewer would — questions, not demands:

## Review Simulation: [Reviewer Name]

### Questions (likely to ask)
**file:line** — [question in their voice]
Fix with: [which agent or direct edit]

### Issues (would flag)
**file:line** — [what's wrong]
Fix with: [which agent]

### What Looks Good
[acknowledge well-done parts]

### Fix Plan
| # | Issue | Fix Agent | Instruction |
|---|-------|-----------|-------------|
| 1 | ... | ... | ... |

Execution order: [dependencies]