Fix Prompt Generation

# Scout LLM Fix Prompt Generation

**Purpose:** Scout-specific action verbs, suppression cases, template fields, and worked example for the `## LLM Fix Prompt` block at the end of every Scout investigation report.
**Read when:** You are writing the `## LLM Fix Prompt` block for a Scout report, choosing an action verb, or deciding whether to suppress.

> Universal authoring rules and prompt structure: `_common/LLM_PROMPT_GENERATION.md`.
> This file documents only Scout-specific verbs, suppression cases, template fields, and an example.

## Contents

- Scout action verbs
- Verb selection heuristic
- Scout-specific suppression cases
- Per-bug fix prompt template (Scout-specific fields)
- Worked example

---

## Scout Action Verbs

Each fix prompt declares one verb at the top of `# Your task`.

| Verb | When to use | Receiving agent / LLM |
|------|-------------|----------------------|
| `FIX` | HIGH confidence, scoped to identified files, no security/concurrency concern | Builder, Claude, Codex |
| `FIX-WITH-TEST` | HIGH confidence + Scout has Radar-quality regression test specs to bundle | Builder + Radar (combined session) |
| `MITIGATE` | Workaround only — root cause is out of scope, blocked, or owned by another team | Builder |
| `INVESTIGATE-FURTHER` | LOW or MEDIUM confidence — receiving LLM must reproduce and verify hypothesis before changing code | Claude / Codex (investigation mode), or Scout (re-entry) |
| `REFACTOR-FIX` | Fix requires structural change beyond a single function; needs architecture review first | Atlas → Builder |

---

## Verb Selection Heuristic