Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/luanpdd/kit-mcp/auditar-marcogit clone --depth 1 https://github.com/luanpdd/kit-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/luanpdd/kit-mcp/auditar-marco)<a href="https://agentmods.dev/commands/luanpdd/kit-mcp/auditar-marco"><img src="https://agentmods.dev/badge/commands/luanpdd/kit-mcp/auditar-marco.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.02343 |
| Opus 5 | $0.00010 | $0.01171 |
| Sonnet 5 | $0.00004 | $0.00469 |
| Haiku 4.5 | $0.00002 | $0.00234 |
Grade A, and why
auditar-marco scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Este comando É o orquestrador. Lê os arquivos VERIFICATION.md existentes (fases já verificadas durante o executar-fase), agrega dívidas técnicas e lacunas adiadas, então invoca o verificador de integração para a fiação entre fases.
<execution_context> @./.claude/framework/workflows/audit-milestone.md </execution_context>
Arquivos principais de planejamento são resolvidos no workflow (init milestone-op) e carregados apenas quando necessário.
Trabalho Concluído: Glob: .planning/phases//-SUMMARY.md Glob: .planning/phases//-VERIFICATION.md
<observability_integration> OMM scoring (v1.9 — INT-FW-04):
Quando workflow.audit_milestone_omm = true (default), o workflow inclui passo OMM scoring:
Skill(skill="framework:auditar-observabilidade")
O comando /auditar-observabilidade invoca o agente omm-auditor que pontua as 5 capacidades (resiliência, qualidade, complexidade, cadência, comportamento) contra o marco anterior. O OMM-REPORT.md gerado é incluído como anexo no MILESTONE-AUDIT.md.
Resultado de regression OMM:
- 0 regressions: audit aprovado
- 1+ regressions, blocking=false: warn explícito; audit aprovado com nota
- 1+ regressions, blocking=true (
workflow.omm_no_regression=true): audit fail → user escolha entre fix lacunas ou aceitar
Skill consultada: observability-maturity-model.
REQ: INT-FW-04. </observability_integration>
<sre_integration> Toil scoring auto-invocação (v1.10 — INT-FW-V2-03):
Quando workflow.audit_milestone_toil = true (default), o workflow inclui passo Toil audit auto-invocação antes do passo de OMM scoring (que já existe via <observability_integration> v1.9 — INT-FW-04):
Skill(skill="framework:auditar-toil")
O comando /auditar-toil invoca o agente toil-auditor que analisa git log recente (≤ 90 dias) + scripts shell em scripts/ + comandos manuais documentados em README/runbooks/.planning/runbooks/ + tarefas repetitivas em .planning/phases/*/SUMMARY.md. O agent classifica candidatos a automação (P0/P1/P2 por esforço × frequência) e produz .planning/TOIL-AUDIT.md na raiz do .planning/. Cap 5 do livro Google SRE (Eliminating Toil) define toil canonicamente: manual + repetitivo + automatizável + tático + sem valor durável + escala linear com tráfego/team.
Loop fechado canônico:
/auditar-marco
↓
Step A: invoca /auditar-toil ← gera .planning/TOIL-AUDIT.md (este patch — INT-FW-V2-03)
↓
Step B: invoca /auditar-observabilidade ← OMM scoring v1.9 (INT-FW-04)
↓
omm-auditor consulta .planning/TOIL-AUDIT.md ← Capacidade 3 — Complexidade / Tech Debt (Phase 39 INT-OBS-02)
↓
OMM-REPORT.md inclui Capacidade 3 score derivado de % toil pelo time
↓
MILESTONE-AUDIT.md inclui OMM-REPORT.md + TOIL-AUDIT.md como anexos
Por que rodar /auditar-toil ANTES de /auditar-observabilidade:
O agent omm-auditor (Capacidade 3 patcheada em Phase 39 / INT-OBS-02) tem regra absoluta:
"score Capacidade 3 > 3 exige TOIL-AUDIT.md fresco ≤ 30 dias com
% toil < 30%"
Se TOIL-AUDIT.md ausente ou stale (> 30d), omm-auditor delega geração via Task(subagent_type=toil-auditor) ad-hoc — duplicação. Auto-invocar /auditar-toil em /auditar-marco evita essa duplicação ao garantir que omm-auditor encontre TOIL-AUDIT.md fresco.
Tabela de score Capacidade 3 (consumida por omm-auditor):
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 180 lines · 19 tokens per session scan A 24072293ca82
auditar-marco is a command published in the GitHub repository luanpdd/kit-mcp (1 stars, last pushed 5d ago), licensed MIT. It adds 19 tokens to every session and 2,343 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.