---
---

# Step 3: Triage

## RULES

- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`

## INSTRUCTIONS

1. **Normalize** findings into a common format. Expected input formats:
   - Adversarial (Blind Hunter): markdown list of descriptions
   - Edge Case Hunter: JSON array with `location`, `trigger_condition`, `guard_snippet`, `potential_consequence` fields
   - Acceptance Auditor: markdown list with title, AC/constraint reference, and evidence

   If a layer's output does not match its expected format, attempt best-effort parsing. Note any parsing issues for the user.

   Convert all to a unified list where each finding has:
   - `id` -- sequential integer
   - `source` -- `blind`, `edge`, `auditor`, or merged sources (e.g., `blind+edge`)
   - `title` -- one-line summary
   - `detail` -- full description
   - `location` -- file and line reference (if available)

2. **Deduplicate.** If two or more findings describe the same issue, merge them into one:
   - Use the most specific finding as the base (prefer edge-case JSON with location over adversarial prose).
   - Append any unique detail, reasoning, or location references from the other finding(s) into the surviving `detail` field.
   - Set `source` to the merged sources (e.g., `blind+edge`).

3. **Read the code before rating.** Before assigning severity, open the source at each finding's location and read enough surrounding code to judge reachability -- call sites, guards, and validation that live outside the diff hunk. Do not rate from the diff hunk alone. Severity reflects the real consequence at a real call site, not the worst theoretical reading.

4. **Assign severity** to each finding by consequence for the artifact's main consumer (software user, document reader, etc).
   Disregard any severity assigned by a reviewing subagent. Review subagents operate under by-design information asymmetry and do not have enough context to set final severity for this workflow.
   - `low` -- none or cosmetic
   - `medium` -- tolerable
   - `high` -- intolerable

5. **Route** each finding into exactly one triage bucket:
   - **decision_needed** -- There is an ambiguous choice that requires human input. The code cannot be correctly patched without knowing the user's intent. Only possible if `{review_mode}` = `"full"`.
   - **patch** -- Code issue that is fixable without human input. The correct fix is unambiguous.
   - **defer** -- Pre-existing issue not caused by the current change. Real but not actionable now.
   - **dismiss** -- Noise, false positive, or handled elsewhere.

   If `{review_mode}` = `"no-spec"` and a finding would otherwise be `decision_needed`, reclassify it as `patch` (if the fix is unambiguous) or `defer` (if not).

6. **Drop** all `dismiss` findings. Record the dismiss count for the summary.

7. If `{failed_layers}` is non-empty, report which layers failed before announcing results. If zero findings remain after dropping dismissed AND `{failed_layers}` is non-empty, warn the user that the review may be incomplete rather than announcing a clean review.

8. If zero findings remain after triage (all rejected or none raised): state "✅ Clean review — all layers passed." (Step 3 already warned if any review layers failed via `{failed_layers}`.)


## NEXT

Read fully and follow `./step-04-present.md`