bug-report
Writes a clear, reproducible defect report from a failing test, log, stack trace or complaint: minimal repro, evidence without secrets, severity, impact and a regression test idea. Use when the user says 'write a bug report', 'file this bug' or 'minimal repro'. Not for hunting new bugs (use bug-hunter).
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- bug-report
- license
- Apache-2.0
- description
- Writes a clear, reproducible defect report from a failing test, log, stack trace or complaint: minimal repro, evidence without secrets, severity, impact and a regression test idea. Use when the user says 'write a bug report', 'file this bug' or 'minimal repro'. Not for hunting new bugs (use bug-hunter).
- argument-hint
- [failing test, log, issue text or finding]
03 SKILL.md 正文
bug-report
Objective
A good bug report lets another person see the bug in minutes, without a meeting. This skill writes such a report, and it checks the facts first.
Context
- Read
.agents/qa-project-context.mdif it exists. Use its environments, components and severity rules. - If it does not exist, collect the facts from the source and the repository. Do not stop the work for the context file.
Discovery questions
- What is the source of the bug, and where does it happen?
- Which tracker and template does the team use? Who sets the priority?
Workflow
- Confirm the bug. Find the requirement or documented behavior that the actual result breaks. If none exists, it can be a missing requirement: write it as a question.
- Reproduce with the shortest steps or the smallest failing test. Remove each step that is not needed. Record the exact data and the environment (version or commit, configuration, client, platform, time zone if it matters).
- Narrow down: what does NOT show the bug (other data, users, versions). If the history is available, name the first version or commit that shows it.
- Evidence: the error message, relevant log lines, the stack trace, a request and a response, the failing test name. Remove secrets and personal data.
- Rate severity (critical: data loss, security, crash of a main path, money; major: wrong result, no workaround; minor: workaround exists; trivial: cosmetic). Propose a priority with a reason, or ask.
- Write the report, and check that a reader who knows the product but not the code can follow it.
Output
Title: <component>: <what is wrong> when <condition>
Environment: <version/commit, configuration, platform>
Severity: <..> Priority (proposed): <..>
Steps to reproduce:
1. ...
Expected result: ... (source: <requirement, contract or document>)
Actual result: ...
Frequency: always | sometimes (<n> of <m> runs)
Does not happen when: ...
Evidence: ...
Impact: <who, how often, what it costs>
Regression test: <the test that must exist after the fix>
Cause hypothesis (optional, labeled): ...
Anti-patterns
- Two bugs in one report.
- A guess about the cause in the steps. Keep it in the labeled hypothesis line.
- Secrets or personal data in the evidence.
- Filing in the tracker without the user's approval.
Done when
- The expected result has a source, and the steps reproduce the bug (or the frequency is stated).
- The evidence is clean of secrets and personal data.
- Severity, proposed priority, impact and a regression test are stated.
Related skills
bug-hunter: finds and proves bugs in code.test-runner: triages failures.quality-postmortem: when the bug escaped to production.exploratory-testing: bugs found in sessions.
04 同仓库更多 Skills
unit-testing
Writes, fixes or improves unit tests for rules, calculations, validation and state logic: arrange-act-assert, exact oracles, boundary sets, doubles only at boundaries, controlled clock and randomness. Use when the user says 'write unit tests', 'TDD this' or 'mock this dependency'. Not for HTTP endpoints (use api-testing).
ui-automation
Writes, fixes or structures browser end-to-end tests for critical journeys: role, label or test ID locators, auto-waiting assertions, API data setup, page objects, auth reuse and traces. Use when the user says 'UI test', 'end-to-end test' or 'our UI tests are brittle'. Not for business rules (use unit-testing).
testcase-reviewer
Reviews automated tests or manual test cases like a senior test architect: scores fidelity, oracles, assertion depth, edge cases and determinism, and finds tests that pass on wrong code. Use when the user says 'review my tests', 'test smells' or 'review AI-written tests'. Not for fault injection (use mutation-probe).
testability-review
Scores how testable a feature, module or design is before tests exist (seams for clock, randomness and network, hidden state, observability) and gives the smallest production change per blocker. Use when the user asks 'is this testable' or 'why is this hard to test'. Not for writing the tests (use unit-testing).
test-strategy
Writes a test strategy for a product, system or team: quality goals, test levels and their share, environments, data, automation scope, non-functional testing, metrics and a roadmap. Use when the user says 'test strategy' or 'QA strategy'. Not for one sprint or release plan (use test-planning).
test-runner
Finds how a project runs its tests, runs the right subset, and triages each failure as product bug, test defect, environment problem or flaky test, with evidence. Use when the user says 'run the tests' or asks 'which tests fail and why'. Not for pipeline design (use ci-cd-integration).