ワンクリックで
triage-issue
Investigate a reported problem, diagnose root cause, and file a Beads issue with a behavior-level TDD fix plan.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Investigate a reported problem, diagnose root cause, and file a Beads issue with a behavior-level TDD fix plan.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | triage-issue |
| description | Investigate a reported problem, diagnose root cause, and file a Beads issue with a behavior-level TDD fix plan. |
Investigate a bug report or broken behavior, find the root cause, and create follow-up work in Beads with a durable RED/GREEN fix plan. Keep the workflow mostly hands-off: ask only for the minimum problem statement, then investigate.
Do not use this skill for broad product planning, feature decomposition, or already-approved implementation work. Use to-prd, to-issues, or tdd instead when those are the actual request.
Get one brief description of the issue from the user. If they have not provided one, ask exactly one clarifying question first:
What's the problem you're seeing?
Do not start with a long questionnaire. Preserve any reported symptoms, expected behavior, reproduction hints, logs, affected environment, and impact, then begin diagnosis.
Before source exploration, read relevant knowledge and repo guidance:
knowledge/_shared.yamlknowledge/repos/<repo>.yaml for the affected repo, when presentrepos/<repo>/AGENTS.md and repos/<repo>/CONTRIBUTING.md when investigating a sub-repobd show <id> when the report is attached to an issueIf a knowledge file is missing, note that briefly and continue. Do not let missing documentation block investigation.
Trace the behavior far enough to identify root cause, not just the symptom. Investigate:
Use a focused Explore subagent only when the codebase area is broad or unfamiliar. For small, local bugs, direct reading and targeted search are enough. Do not delegate merely to add ceremony.
Keep private diagnostic notes as needed, but keep the final issue durable. The issue should describe behaviors, contracts, and outcomes rather than fragile file paths, line numbers, or implementation details that could become stale after a refactor.
Before filing, decide:
If root cause is not known, file an investigation task rather than pretending to know the fix. State the leading hypothesis and the evidence needed to confirm it.
Write an ordered list of RED/GREEN cycles. Each cycle is one vertical slice:
RED: Add one failing test that captures an observable broken behavior.
GREEN: Make the smallest code change that passes that test.
Rules:
Before creating a new Beads issue or changing an existing issue's priority, read and apply:
.claude/skills/beads-priority-assignment/SKILL.md
Set an explicit priority on every created issue when the local bd CLI supports priority. If no urgency, impact, or deferral-risk signal exists, use the harness default P2 / 2 / medium.
Use Beads by default. Do not create a GitHub issue unless the user explicitly asks for GitHub or the repository's documented workflow requires it.
For a new issue:
bd create --type task --title "<durable behavior title>" --description "<issue body markdown>" --priority <value> --repo .
bd comments add <new-id> "worklog: triaged root cause; TDD fix plan added."
For an existing issue:
bd comments add <id> "worklog: triage result - <one-line root cause and TDD plan summary>."
bd update <id> --priority <value>
Use this issue body shape:
Describe the durable behavior:
Summarize the diagnosis:
Avoid file paths, line numbers, and implementation-specific diffs in this section. Mention modules only at the level a maintainer would still understand after a refactor.
RED: Write a behavior test proving <expected observable outcome>.
GREEN: Implement the smallest change that makes that behavior pass.
RED: Write a behavior test proving <next important edge case or regression guard>.
GREEN: Extend the implementation only as needed for that behavior.
REFACTOR: Clean up duplication or naming after all planned tests pass, then rerun the relevant quality gates.
After creating or updating the issue, report the Beads ID, chosen priority, and a one-line root-cause summary to the user.
Run red-green-refactor in vertical slices, emphasizing behavior-first tests and harness quality gates.
Break a plan/PRD into independent vertical-slice Beads issues with explicit dependencies.
Synthesize current context into a PRD and file it as a Beads feature/epic issue.
Phase 2 of the Forge pipeline. Turn a research document into an implementation plan that is built around Beads issues and TDD. Reads plans/research/<slug>.md, designs the build, sets up red-green-refactor test strategy using the tdd skill, materializes a Beads epic/feature/tasks with dependencies and priorities, and defines demo/test checkpoints. Use when starting /forge-plan or the plan step of /forgemaster.
Fact-check an implementation plan against the real codebase and first-party documentation, flagging inaccuracies with what / why / evidence and a severity tier. Use when the user wants a plan reviewed or validated for accuracy, runs /review-plan, or wants a plan checked against real code before implementing. Accepts a Beads issue, a filesystem path, or a plan pasted into chat.
Phase 3 of the Forge pipeline. Execute the plan's Beads tasks with red-green-refactor TDD, committing per checkpoint and pausing at each natural stopping point so the user can run a live demo or test and see the work themselves. Use when starting /forge-implement or the implement step of /forgemaster.