Bug TriageSAFE

Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: bug-triage
description: "Re-evaluate open bugs — priority vs severity, assign to sprints, surface systemic trends. Run when the count grows."
argument-hint: "[sprint | full | trend]"
user-invocable: true
allowed-tools: Read, Glob, Grep, Write, Edit, Bash(bash "*/.claude/skills/bug-triage/../../hooks/yaml-helper.sh" resolve_config *)
model: sonnet
---

!`bash "${CLAUDE_SKILL_DIR}/../../hooks/yaml-helper.sh" resolve_config --keys automation`

**Automation mode**: Resolve `modes.automation` (`project.local.yaml` →
`project.yaml` → default `collaborative`). Every `AskUserQuestion` call and
every file write follows `.claude/docs/automation-modes.md`
(collaborative asks always · guided major-only · autonomous logs and proceeds;
`automation_always_ask` categories always prompt).

# Bug Triage

This skill processes the open bug backlog into a prioritised, sprint-assigned
action list. It distinguishes between **severity** (how bad is the impact?) and
**priority** (how urgently must we fix it?), detects systemic trends, and
ensures no critical bug is lost between sprints.

**Output:** `production/qa/bug-triage-[date].md`

**When to run:**
- Sprint start — assign open bugs to the new sprint or backlog
- After `/team-qa` completes and new bugs have been filed
- When the bug count crosses 10+ open items

---

## 1. Parse Arguments

**Modes:**
- `/bug-triage sprint` — triage against the current sprint; assign fixable bugs
  to the sprint backlog; defer the rest
- `/bug-triage full` — full triage of all bugs regardless of sprint scope
- `/bug-triage trend` — trend analysis only (no assignment); read-only report
- No argument — run sprint mode if a current sprint exists, else full mode

---

## 2. Load Bug Backlog

### Step 2a — Discover bug files

Glob for bug reports in priority order:
1. `production/qa/bugs/*.md` — individual bug report files (preferred format)
2. `production/qa/bugs.md` — single consolidated bug log (fallback)
3. Any `production/qa/qa-plan-*.md` "Bugs Found" table (last resort)

If no bug files found:
> "No bug files found in `production/qa/bugs/`. If bugs are tracked in a
> different location, adjust the glob pattern. If no bugs exist yet, there is
> nothing to triage."

Stop and report. Do not proceed if no bugs exist. Verdict: **COMPLETE** — no bug files in `production/qa/bugs/`; nothing to triage.

**In `trend` mode, do not read full bug bodies.** Trend metrics (volume, severity
mix, by-system, by-date) are computable from the header fields alone:
```
Grep pattern="\*\*(Severity|Priority|Status|System|Category|Reported)\*\*" glob="production/qa/bugs/*.md" output_mode="content"
```
(Bug-report fields are bolded — `**Severity**:`, `- **System**:` — so match the
`**field**` form, not a bare line-start `Field:`.)
Full bug bodies are needed only for the priority-vs-severity **re-evaluation** in
`sprint`/`full` modes; `trend` is a read-only report and skips it. (The one
deviation check that needs a story's status — "bug filed against a Complete
story" — is a targeted story-status grep either way, not a bug-body read.)

### Step 2b — Load sprint context

Read the most recently modified file in `production/sprints/` to understand:
- Current sprint number / name
- Stories in scope (for assignment target)
- Sprint capacity constraints (if noted)

If no sprint file exists: note "No sprint plan found — assigning to backlog only."

### Step 2c — Load severity reference

Read `.claude/docs/coding-standards.md` for severity/priority definitions if they
exist. If they do not exist, use the standard definitions in Step 3.

---

## 3. Classify Each Bug

For each bug, extract or infer:

### Severity (impact of the bug)

| Severity | Definition |
|----------|-----------|
| **S1 — Critical** | Game crashes, data loss, or complete feature failure. Cannot proceed past this point. |
| **S2 — High** | Major feature broken but game is still playable. Significant wrong behaviour. |
| **S3 — Medium** | Feature degraded but a workaround exists. Minor wrong behaviour. |
| **S4 — Low** | Visual glitch, cosmetic issue, typo. No gameplay impact. |

### Priority (urgency of the fix)

| Priority | Definition |
|----------|-----------|
| **P1 — Fix this sprint** | Blocks QA, blocks release, or is regression from last sprint |
| **P2 — Fix soon** | Should be resolved before the next major milestone |
| **P3 — Backlog** | Would be good to fix, but no active blocking impact |
| **P4 — Won't fix / Deferred** | Accepted risk or out of scope for current product scope |

### Assignment

For each P1/P2 bug in `sprint` mode:
- Identify which story or epic the fix belongs to
- Check whether the current sprint has remaining capacity
- If capacity exists: assign to sprint (`Sprint: [current]`)
- If capacity is full: flag as `Priority overflow — consider pulling from sprint`

For `full` mode: assign all P1 to current sprint, P2 to next sprint estimate,
P3+ to backlog.

### Deviation check

Flag bugs that suggest **systematic problems**:
- 3+ bugs from the same system in the same sprint → "Potential design or
  implementation quality issue in [system]"
- 2+ S1/S2 bugs in the same story → "Story may need to be reopened and
  re-reviewed before shipping"
- Bug filed against a story marked Complete → "Regression in completed story —
  story should be re-opened in sprint tracking"

---

## 4. Trend Analysis

After classifying all bugs, generate trend metrics:

### Volume trends
- Total open bugs: [N]
- Opened this sprint: [N]
- Closed this sprint: [N]
- Net change: [+N / -N]

### System hot spots
- Which system has the most open bugs?
- Which system has the highest S1/S2 ratio?

### Age analysis
- How many bugs are older than 2 sprints?
- Are any S1/S2 bugs un-assigned (sprint = none)?

### Regression indicator
- Any bugs filed against previously-completed stories?
- Count: [N] regression bugs (story reopened implied)

---

## 5. Generate Triage Report

```markdown
# Bug Triage Report

> **Date**: [date]
> **Mode**: [spri

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.