tech-debt
Skill4 starsChanged 46 days agoMITView on GitHub
Inventories technical debt — outdated dependencies vs the stack reference, TODO/FIXME, skipped tests, lint suppressions, ADR drift, missing docs, security/perf shortcuts — scores it by impact/effort and proposes stories; report in docs/ops/tech-debt-<date>.md on approval. Use before sprint planning or on an adopted project, for 'how much debt', 'what should we clean up'.
36 other projects use a version of this.
Adopted by
Most recently taken up by
- nisarth/claude-skills-storetheir own version, first seen Nov 8, 2026
- dirien/yet-another-agent-harnesstheir own version, first seen Nov 8, 2026
- y-miyazaki/configtheir own version, first seen Nov 6, 2026
- striderZA/OpenCodeGameStudiostheir own version, first seen Nov 4, 2026
- dpleshakov/tfsclitheir own version, first seen Oct 20, 2026
- Donchitos/Claude-Code-Game-Studiostheir own version, first seen Oct 19, 2026
Used with
Projects with it also use playtest-report, soak-test and 8 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.
CoralGameSkillplaytest-report
Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.
Also in 12 projects that use this
lilangMaxSkillsoak-test
Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.
Also in 9 projects that use this
lilangMaxSkillteam-live-ops
Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.
Also in 9 projects that use this
CoralGameSkillteam-polish
Orchestrate the polish team: coordinates performance-analyst, technical-artist, sound-designer, and qa-tester to optimize, polish, and harden a feature or area for release quality.
Also in 14 projects that use this
- +3 more from lilangMax
- +3 more from CoralGame
Other versions36 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 36
- anthropics/knowledge-work-plugins★ 35ksame name
- Donchitos/Claude-Code-Game-Studios★ 27ksame name
- vm0-ai/vm0★ 1.2ksame name
- pixel-cellar/Claude-Code-Game-Studios★ 324same name
- striderZA/OpenCodeGameStudios★ 118same name
- tranhieutt/software_development_department★ 71same name
- cenconq25/claude-code-app-studio★ 40same name
- drafthq/draft★ 40same name
- fagemx/edda★ 36same name
- CoralGame/Claude-Code-Game-Studios★ 19same name
26 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 repository52 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- testing/CLAUDE.mdCLAUDE.md
- a11y-auditSkill
- adoptSkill
- api-contractSkill
- architecture-decisionSkill
- architecture-reviewSkill
- backlogSkill
- brainstormSkill
- changelogSkill
- code-reviewSkill
- collaboration-protocolSkill
- create-storiesSkill
- data-modelSkill
- dependency-auditSkill
- deploySkill
- design-systemSkill
- dev-storySkill
- docsSkill
- feature-specSkill
- game-conceptSkill
- hardenSkill
- helpSkill
- hotfixSkill
- impactSkill
- incidentSkill
- initSkill
- migrateSkill
- pentestSkill
- perf-auditSkill
- product-specSkill
- qa-planSkill
- refactorSkill
- release-checklistSkill
- retrospectiveSkill
- security-auditSkill
- setup-stackSkill
- skill-improveSkill
- skill-testSkill
- sprint-planSkill
- sprint-statusSkill
- stack-updateSkill
- startSkill
- story-doneSkill
- team-featureSkill
- team-gameSkill
- team-releaseSkill
- team-securitySkill
- test-setupSkill
- threat-modelSkill
- updateSkill
- ux-specSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 29 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 35
After the "write" answer: touch .claude/.write-consent (rule 7 — the consent-guard hook checks the marker), then write.
- Very specific: line 40
When HEAD is a story branch, name it and ask one AskUserQuestion: switch to the default branch and commit there (Recommended — a …
“Common” is measured: kept by at least 6 in 10 of the 29 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
Tech Debt
Reply in the project conversation language (CLAUDE.md → Language); code, identifiers, paths and commit messages stay in English.
References stack-reference/index.md (and php.md / go.md for the code-shape numbers). This skill delegates nothing: the inventory is command and grep output, the scoring is the director's own, so no Task — the stories it proposes are /create-stories' work, not a subagent's. In the commands below, <hooks> is .claude/hooks/ in copy mode and ${CLAUDE_PLUGIN_ROOT}/hooks/ in plugin mode.
Phase 1: Collect (Bash + Grep)
- Scope: the argument names an area (a package, a directory, a layer);
fullor no argument scans the whole repository. An area limits every search below to that subset. - No code to scan → say so and report an empty inventory; nothing is written.
- Collect:
- dependencies:
go list -m -u all/composer outdated/pnpm outdatedvsstack-reference/index.md; - TODOs by
rules/comments.md—git grep -nE 'TODO|FIXME|HACK'split into: with an id whose card is Done or missing (grep -c '<ID>' production/roadmap.md production/backlog.md, the line's[x]) → a row each; with an open id → listed by story; without an id → one row with the count (each is a/backlog add); - comment history —
git grep -nE '^\s*(//|#|\*) .*(\b(S|OPS|ARCH|SEC)-[0-9]+\b|pre-S-|used to|previously)' -- ':!*_test.go' ':!*Test.php' ':!*.spec.ts' ':!*.test.ts' | grep -vE '(TODO|FIXME|HACK)\((S|I)-[0-9]+\)' | cut -d: -f1 | sort | uniq -c(the well-formed TODOs of the previous item are not history) summed per package, and for Gogo doc -all ./... 2>/dev/null | grep -cE '\b(S|ARCH|OPS)-[0-9]+\b|used to'(what the rendered API documentation carries) — a row per package over 20 lines, proposed storychore(<package>): comments per rules/comments.md(never a drive-by in a feature story); - skipped tests
skip|xit|@group skip|t.Skip; lint suppressionseslint-disable|@psalm-suppress|nolint; - ADRs in Proposed older than 30 days; feature specs without stories; Done stories without tests; missing runbooks;
- open findings from the last audits (security/perf/a11y).
- dependencies:
- Code shape is measured, not felt — each item a row with its number:
- PHP: classes over 400 lines, a PHPUnit major out of support,
sleep()in tests, exception messages asserted, framework namespaces imported undersrc/Domain/src/Applicationin alayeredproject, DDL outside migrations, aphp_static_analysislevel below the recorded one. Proposed story:/refactor <namespace>//refactor layout//refactor tests//refactor framework. - Go: packages over 1 500 lines, files over 500, a
.golangci.ymlin v1 format,time.Sleepin tests, error strings compared, ago_architecture: layeredproject whose tree ordepguardblock does not match. Proposed story:/refactor <package>//refactor layout//refactor tests, rather than a rewrite.
- PHP: classes over 400 lines, a PHPUnit major out of support,
Phase 2: Score
Table "debt → impact (security/velocity/risk) → effort → priority → proposed story". A dependency a major behind the stack reference gets /stack-update as its proposed action.
Phase 3: Report
- Show the table in the chat.
- "May I write
docs/ops/tech-debt-<date>.mdand add the top 5 to the roadmap?" as oneAskUserQuestion: report and roadmap (Recommended) · report only · not now. - After the "write" answer:
touch .claude/.write-consent(rule 7 — the consent-guard hook checks the marker), then write.
Phase 4: Commit (documents lane)
Right after the write, one commit gate (rule 7 (4), .claude/docs/git-workflow.md § Documents): docs: tech debt <date>, staging exactly the written files — docs/ops/tech-debt-<date>.md and production/roadmap.md when the top 5 were added to it. Record the gate before asking — <hooks>session-state.sh set Gate "/tech-debt Phase 4: commit?" — and clear it after the answer (<hooks>session-state.sh set Gate "—").
- On the default branch when no story work is in progress: one
AskUserQuestion— commit (Recommended) · leave uncommitted. - When HEAD is a story branch, name it and ask one
AskUserQuestion: switch to the default branch and commit there (Recommended — a pipeline-wide document) · commit here (the document belongs to this story) · leave uncommitted. - The inventory is read-only: no code changes exist to stage; a debt item fixed on the way would be a story or the chore lane (git-workflow.md § Chore / infra), never part of the
docs:commit.
Nothing is committed without the answer.
Verdict: COMPLETE (N items, M critical). Next step — one AskUserQuestion: /create-stories for critical items (Recommended when any) · /refactor --dry-run for code-shape items · /stack-update for outdated majors · stop here.
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.

