bug-report
Skill40 starsChanged 7 months agoMITView on GitHub
Author a structured bug report from a description, OR analyze code/logs to derive likely bugs. Captures reproduction steps, severity, device matrix, expected vs actual, and attachments. Use after a manual QA failure or when the user describes a defect.
33 other projects use a version of this.
Adopted by
Most recently taken up by
- alexpalexpinne/qa-toolkittheir own version, first seen Nov 8, 2026
- github/plugin-marketplace-startertheir own version, first seen Nov 8, 2026
- craigcossairt/trellistheir own version, first seen Nov 6, 2026
- striderZA/OpenCodeGameStudiostheir own version, first seen Nov 4, 2026
- Suryanandx/qvac-uitheir own version, first seen Nov 1, 2026
- oasm-platform/open-asmtheir own version, first seen Oct 29, 2026
Used with
Projects with it also use setup-engine, team-audio and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
CoralGameSkillsetup-engine
Configure the project's game engine and version. Pins the engine in CLAUDE.md, detects knowledge gaps, and populates engine reference docs via WebSearch when the version is beyond the LLM's training data.
Also in 13 projects that use this
CoralGameSkillteam-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
Also in 13 projects that use this
- +3 more from CoralGame
Other versions33 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 33
- Donchitos/Claude-Code-Game-Studios★ 27ksame name
- pipecat-ai/gradient-bang★ 420same name
- pixel-cellar/Claude-Code-Game-Studios★ 324same name
- oasm-platform/open-asm★ 272same name
- striderZA/OpenCodeGameStudios★ 118same name
- tranhieutt/software_development_department★ 71same name
- CoralGame/Claude-Code-Game-Studios★ 19same name
- composio-community/support-skills★ 18same name
- dawn840705/codex-code-studios★ 17same name
- akovalion/paranoid-qa★ 16same name
23 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository77 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- design/CLAUDE.mdCLAUDE.md
- docs/CLAUDE.mdCLAUDE.md
- src/CLAUDE.mdCLAUDE.md
- tests/CLAUDE.mdCLAUDE.md
- adoptSkill
- architecture-decisionSkill
- architecture-reviewSkill
- asset-auditSkill
- asset-specSkill
- balance-checkSkill
- brainstormSkill
- bug-triageSkill
- changelogSkill
- code-reviewSkill
- consistency-checkSkill
- content-auditSkill
- create-architectureSkill
- create-control-manifestSkill
- create-epicsSkill
- create-storiesSkill
- day-one-patchSkill
- design-bibleSkill
- design-reviewSkill
- design-systemSkill
- dev-storySkill
- estimateSkill
- extractSkill
- gate-checkSkill
- helpSkill
- hotfixSkill
- launch-checklistSkill
- localizeSkill
- map-systemsSkill
- milestone-reviewSkill
- onboardSkill
- patch-notesSkill
- perf-profileSkill
- prd-reviewSkill
- project-stage-detectSkill
- propagate-design-changeSkill
- prototypeSkill
- qa-planSkill
- quick-designSkill
- regression-suiteSkill
- release-checklistSkill
- retrospectiveSkill
- reverse-documentSkill
- review-all-prdsSkill
- scope-checkSkill
- security-auditSkill
- setup-frameworkSkill
- skill-improveSkill
- skill-testSkill
- smoke-checkSkill
- soak-testSkill
- sprint-planSkill
- sprint-statusSkill
- startSkill
- story-doneSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 6 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 31 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 19
If --from-description (or no flag): collect details from the user via AskUserQuestion.
- Very specific: line 22
If --from-code=<path>: read the specified file(s) and look for likely bugs (unsafe optionals, race conditions, leaked subscriptions, etc.).
- Very specific: line 25
If --from-crashlog=<path>: read the crash log, identify the failing stack frame, and reverse-engineer steps.
- Distinctive: line 52
Keep collected answers in working memory and assemble them in Phase 4.
- Distinctive: line 58
Read the target file(s). Check for these patterns:
- Distinctive: line 80
Read the crash log. Extract:
- Distinctive: line 88
Search the codebase for the failing symbol or file.
- Distinctive: line 91
If the log lacks an app-level frame (pure framework crash), surface that and suggest enabling symbolication.
- Very specific: line 168
Ask: "May I write this to production/qa/bugs/BUG-[NNN]-[slug].md?"
- Very specific: line 186
If the bug is suspected to break a regression test target, propose: "Add this bug to the regression suite via /regression-suite …
- Distinctive: line 189
If the bug stems from a story that recently closed, propose linking it in that story's Closing Notes.
- Very specific: line 192
If a hotfix is warranted, propose /hotfix BUG-NNN.
“Common” is measured: kept by at least 6 in 10 of the 31 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Bug Report
Turn a vague observation into a fileable, actionable defect record. Three input modes — user description, target code/files, or a crash log.
Phase 1: Pick the Mode
If --from-description (or no flag): collect details from the user via
AskUserQuestion.
If --from-code=<path>: read the specified file(s) and look for likely
bugs (unsafe optionals, race conditions, leaked subscriptions, etc.).
If --from-crashlog=<path>: read the crash log, identify the failing
stack frame, and reverse-engineer steps.
Phase 2A: From User Description
Walk through these prompts via AskUserQuestion in order:
- Title — one line, action + outcome ("Sign-in fails after biometric prompt cancel").
- Severity — S1 (crash, data loss, blocks release), S2 (major feature broken, no workaround), S3 (minor feature broken, workaround exists), S4 (cosmetic).
- Affected platforms — iOS / Android / both.
- Device(s) reproduced on — model + OS version. List all confirmed.
- Build version + commit SHA — the build under which the bug was seen.
- Reproduction steps — numbered, unambiguous. Each step should be verifiable on a fresh install.
- Expected result.
- Actual result.
- Frequency — every time / often (>50%) / sometimes / once.
- Attachments — screenshots, screen recordings, crash logs, charles captures, video file paths.
- Workaround — if any.
Keep collected answers in working memory and assemble them in Phase 4.
Phase 2B: From Code
Read the target file(s). Check for these patterns:
- Force-unwrapped optionals on user input.
- Unchecked array index access.
- Async work without cancellation that captures view state.
- Missing offline / error / loading branches.
- Non-idempotent network mutations (no idempotency key).
- Missing feature-flag fallbacks.
- Leaked subscriptions / observers (RN useEffect missing cleanup, Combine cancellables not stored, Flutter StreamSubscription not cancelled, Android Job not cancelled).
- Hardcoded keys, URLs, or environment names.
- Race conditions in
Promise.allordering. - Memory: closures capturing strong
selfin iOS, missing[weak self]. - Deep links accepting unsanitized parameters.
For each finding, propose a bug report with synthesized reproduction.
Phase 2C: From Crash Log
Read the crash log. Extract:
- Crash type (NSException, SIGSEGV, fatal Dart error, ANR, OutOfMemory).
- Top frame in the app's own code.
- The thread (main vs background).
- The OS version, device, build.
- The date and time.
Search the codebase for the failing symbol or file. Read the surrounding code. Hypothesize the trigger condition. Propose reproduction steps.
If the log lacks an app-level frame (pure framework crash), surface that and suggest enabling symbolication.
Phase 3: Triage
For every assembled report, derive:
- Severity — apply S1-S4 rules. Confirm with the user if ambiguous.
- Frequency — affects priority.
- Priority — derived from severity + frequency + business impact: P0 (drop everything), P1 (current sprint), P2 (next sprint), P3 (backlog).
Surface the proposed severity/priority via AskUserQuestion and let the user override.
Phase 4: Render the Report
Use this template:
# BUG-[NNN]: [Title]
**Filed**: [date] by [author]
**Severity**: S[1-4]
**Priority**: P[0-3]
**Status**: Open
**Affected platforms**: iOS / Android / both
**Affected build**: [version + commit]
**Frequency**: [every time / often / sometimes / once]
## Devices Reproduced
- [Model + OS]
- [Model + OS]
## Reproduction Steps
1. ...
2. ...
## Expected
[what should happen]
## Actual
[what happens instead]
## Workaround
[if any]
## Attachments
- [path or link]
## Suspected Area
[code path, module, recent commit, related ADR/PRD]
## Related
- Story: [path or ID]
- ADR: [ID]
- PRD: [path]
- Similar bugs: [list]
## Notes
[free text — investigation log starts here]Number the bug as the next available BUG-NNN by globbing
production/qa/bugs/BUG-*.md.
Slug naming: BUG-NNN-[short-kebab-slug].md.
Phase 5: Save
Ask: "May I write this to production/qa/bugs/BUG-[NNN]-[slug].md?"
On approval, write the file. Append to
production/session-state/active.md:
## Bug Filed — [date]
- ID: BUG-NNN
- Title: [title]
- Severity / Priority: S[N] / P[N]
- Path: production/qa/bugs/BUG-[NNN]-[slug].md
- Next: /bug-triage to schedule, or hand to engineerPhase 6: Cross-Link
If the bug is suspected to break a regression test target, propose:
"Add this bug to the regression suite via /regression-suite --add-bug=BUG-NNN."
If the bug stems from a story that recently closed, propose linking it in that story's Closing Notes.
If a hotfix is warranted, propose /hotfix BUG-NNN.
Quality Gates / PASS-FAIL
A bug report passes if it has:
- A non-empty title that describes outcome, not feeling.
- Reproduction steps that another engineer can follow on a fresh build.
- Expected vs actual cleanly separated.
- A severity AND priority assigned.
- A device list with at least one entry.
- A build version + commit SHA OR a clear note "build version unknown".
Reject and re-prompt for any missing field.
Examples
Example 1 — User description: Tester reports "the sign-in screen sometimes hangs". Skill walks through every prompt, derives S2/P1, writes BUG-067 with 5-step repro, lists iPhone 13 + iOS 17.5 + iPhone SE + iOS 16.7.
Example 2 — From crash log:
User pastes a .crash path. Skill extracts the top frame
(PaymentReducer.applyPromo), reads that file, sees an unguarded
force-unwrap on a server response field, drafts BUG-068 with
hypothesized repro: "Apply promo code with empty discount payload".
Next Steps
- High-severity bug ->
/hotfix BUG-NNNto launch the emergency flow. - Otherwise ->
/bug-triageat sprint start to schedule. - For S1/S2 closes ->
/regression-suite --add-bug=BUG-NNN.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

