harness-audit
Skill5 starsChanged 25 days agoMITView on GitHub
Audita a doutrina do keelson (CLAUDE.md, commands/, agents/, skills/, guidelines/, conventions/, templates/) em duas camadas — ponteiros mecânicos via scripts/check-refs.sh e juízes duplos cegos de poda com a régua 4.160 e plants de controle (4.186). Report-only: aplicar corte é leva própria. Ativar quando o Diretor pedir auditoria do harness, poda/simplificação da doutrina, ou revisão de redundância/utilidade das instruções. Ferramenta do MANTENEDOR (fora do pacote — decisão 4.182).
26 other projects use a version of this.
Adopted by
Most recently taken up by
- hoshiori-dev/harness.mdtheir own version, first seen Nov 16, 2026
- lumlaolao/spike-evaltheir own version, first seen Nov 16, 2026
- gutohuida/AgentWeavetheir own version, first seen Nov 15, 2026
- 5uck1ess/devkittheir own version, first seen Nov 15, 2026
- venhdev/skillstheir own version, first seen Nov 8, 2026
- DaeSeokSong/oh-my-ouroborostheir own version, first seen Nov 1, 2026
Used with
Projects with it also use project-init, autopilot 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.
u9401066Skillproject-init
Initialize new projects using this template. Triggers: init, new, 新專案, 初始化, create project, 建立專案, bootstrap, scaffold project.
Also in 3 projects that use this
ibahgatSkillautopilot
Run autonomous multi-stage execution loops until acceptance criteria pass or blockers are reached.
Also in 3 projects that use this
Microsoft AzureSkilltest
Also in 3 projects that use this
nicolasestremSkillbrainstorm
Structured idea generation and solution exploration with systematic analysis.
Also in 3 projects that use this
GoogleSkilltdd
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Also in 3 projects that use this
Other versions26 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 26
- AnastasiyaW/claude-code-config★ 214same name
- AnastasiyaW/codex-claude-code-config★ 214same name
- Kanevry/session-orchestrator★ 53same name
- fabianmartinelli-fm/harness-audit★ 7same name
- fmslutions/harness-audit★ 7same name
- 5uck1ess/devkit★ 4same name
- gutohuida/AgentWeave★ 4same name
- DaeSeokSong/oh-my-ouroboros★ 2same name
- lumlaolao/spike-eval★ 2same name
- Sagargupta16/agent-brain★ 2same name
16 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 repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- field-intakeSkill
- idea-forgeSkill
- pr-reviewSkill
- skill-standardsSkill
- plan-validatorSkill
- screen-verifySkill
- spec-validatorSkill
- statusSkill
- task-validatorSkill
The file
Files
Skill: harness-audit
Auditoria interna do harness do keelson. Esta skill é um protocolo de rodada — a régua
de poda tem dono (CLAUDE.md §Registro e governança, bullet 4.160) e o fato mecânico tem
motor (scripts/check-refs.sh); ela só impõe a sequência e as cegueiras. Report-only:
a rodada termina no relatório; aplicar corte é leva própria com análise de impacto (4.181).
Rodada
- Fato mecânico primeiro (4.82). Rode
bash scripts/check-refs.she cite a saída como fato — ponteiro quebrado é defeito objetivo, não vai para os juízes. - Baralho de superfícies. Enumere por
git ls-files(nuncafind .— worktrees sob.claude/são cópia integral do repo e contaminariam o juízo de redundância):CLAUDE.md·commands/*.md·agents/*.md·skills/**/*.md·guidelines/**/*.md·docs/_meta/conventions/*.md·templates/*.md. Copie as superfícies para.harness-eval/runs/<AAAA-MM-DD>-interno/deck/e anote quais perfis têmreviewed: true(poda aplicada neles exigirá re-olhada humana). - Plants só na cópia (4.186). Plante no deck 2 controles SLIM (teoria genérica sem
delta do repo) e 1 controle KEEP (política com consequência concreta, nunca cópia
verbatim de superfície existente); gabarito em
plants-key.jsonno run dir. Invariante: a árvore versionada nunca recebe plant — o fecho da rodada confirmagit statuslimpo e o relatório o declara. - Juízes duplos, segundo cego. Dois subagents pontuam cada superfície do deck; o
Judge2 não lê o score do Judge1 nem o
plants-key.json. Régua primária: os três testes da 4.160 (no-op · sedimento · leading word — leia no dono, não reproduza de memória). Lentes complementares: redundância (custo de redescoberta 0–3; custo ≥ 2 nunca é redundante) e utilidade (deletar muda o comportamento?). Dono único conta: frase que duplica regra de outro dono é achado mesmo sendo útil. Ambos registrammodel: <id>— utilidade é sensível a modelo. - Merge com gates. Veredito por superfície:
KEEP·MIXED(com KEEP/CUT nomeados e acionáveis por seção) ·SLIM·HOLD. Desacordo ou incerteza → sempreHOLD. Juiz que errar mais de 1 família de plant invalida a banda de corte da rodada (vira tudoHOLD; o relatório diz o porquê). Corte grande sugerido → recomendar re-julgamento em segundo modelo antes de qualquer leva de aplicação.
Entrega
Relatório no run dir + resumo ao Diretor: saída do check-refs · contagens por veredito ·
achados acionáveis (SLIM/MIXED com evidência) · status dos plants · confirmação de árvore
limpa. Aplicação, se o Diretor pedir: leva própria com 4.181 (impact-scout quando tocar a
malha), escalação com proposta + default, e re-olhada humana para perfil reviewed: true.
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.

