From ship
Fixes small bugs and minor issues with root cause analysis and TDD verification. Supports finding IDs from audit snapshots and standard bug descriptions.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ship:fix [bug or issue description]When to use
バグ修正, 直して, 修正して, fix bug, 不具合
[bug or issue description]opusThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Rapidly fix small bugs with root cause analysis and TDD verification.
Rapidly fix small bugs with root cause analysis and TDD verification.
$ARGUMENTS holds the bug description or a finding ID from /audit snapshot (e.g., RC-001, SEC-003). Scope is limited to small, well-understood issues of 1-3 files. The $ARGUMENTS pattern routes the mode.
| Pattern | Mode | Action |
|---|---|---|
/^[A-Z]+-[0-9]+$/ | Finding ID Resolution | Read the latest snapshot from ${CLAUDE_SKILL_DIR}/../../workspace/history/ and find the ID match in findings[]. Carry severity / fix_type / root cause, skip Outcome Anchor and Build Check, enter Triage. If absent, present error + suggest Standard Flow |
| empty | Fix Prompt | Ask via AskUserQuestion for Fix type (Bug fix / Error message / Test failure) and Description (free text via Other), then execute |
| otherwise | Standard Flow | Treat as a bug description and run from Outcome Anchor |
| Type | Target | Purpose |
|---|---|---|
| Skill | use-context-root-cause-analysis | 5 Whys for non-obvious bugs |
| Agent | generator-test | Regression test from symptom + repro steps |
| Agent | resolver-build | TypeScript or build error triage |
| Reference | ${CLAUDE_SKILL_DIR}/references/defense-in-depth.md | Multi-layer validation for Recurring / Systematic |
Read .claude/OUTCOME.md before Build Check. If absent, generate the stub via /outcome. Confirm the bug or fix lives inside the outcome state. If outside, § Escalation.
Detect the build command from package.json or project config and run it.
| Result | Action |
|---|---|
| Build errors | Task with subagent_type: resolver-build, then END |
| No errors | Continue to Triage |
Obvious skips both RCA and regression test generation, so it is limited to findings with low misfix risk.
| Input | Condition | Path |
|---|---|---|
| Bug desc | Single location identified + 1-3 line fix + no similar pattern | Obvious |
| Bug desc | Intermittent, multiple repro conditions, or unknown root cause | Non-obvious |
| Finding ID | fix_type: auto and severity low / med | Obvious |
| Finding ID | Otherwise (critical / high, or not auto) | Non-obvious |
Skill("use-context-root-cause-analysis"). If via Finding ID, pass the snapshot root cause as the 5 Whys starting point. Output Symptom / Root cause / Pattern.Task(subagent_type: generator-test) for regression test (pass symptom + repro only)${CLAUDE_SKILL_DIR}/references/defense-in-depth.mdBranch on objective triggers, not confidence self-assessment. Do not attempt fix #4 without escalating.
| Trigger | Action |
|---|---|
| RCA cannot identify root cause | Escalate to /research |
| Tests still fail after fix | Re-analyze root cause. After 3 failures, escalate to /research |
| Multi-file impact (4+ files) | Delegate to /code |
| New feature scope | Delegate to /think |
| Pattern = Systematic | Escalate to /research |
| Fix outside OUTCOME.md scope | Confirm with user; redefine Non-goals or delegate to /code |
| Error | Action |
|---|---|
| resolver-build fails | Present error, ask user for guidance |
| generator-test timeout | Skip regression test, proceed with fix |
| Check | Required |
|---|---|
| Root cause identified (Non-obvious path) | Yes |
| All tests pass | Yes |
| Pattern field recorded from RCA | Yes (Non-obvious path) |
| defense-in-depth applied if needed | Yes (Recurring / Systematic only) |
| Re-audit suggested if via Finding ID | Yes (Finding ID path) |
npx claudepluginhub thkt/dotclaude --plugin shipRoot cause based one-shot bug fix. Runs a full investigation pipeline: debugger diagnosis, gap analysis, requirements generation, execution, and verification. Includes QA suggestions after successful fix.
Enforces a structured bug-fix workflow: reproduce, isolate, apply smallest fix, and verify. Helps investigate errors, debug crashes, and make failing tests pass.