Borrowing it
Nothing to install: this file belongs to DuqueOM/ML-MLOps-Portfolio. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/DuqueOM/ML-MLOps-Portfolio/main/.devin/skills/rule-audit/SKILL.mdgit clone --depth 1 https://github.com/DuqueOM/ML-MLOps-PortfolioWrote 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/skills/duqueom/ml-mlops-portfolio/rule-audit)<a href="https://agentmods.dev/skills/duqueom/ml-mlops-portfolio/rule-audit"><img src="https://agentmods.dev/badge/skills/duqueom/ml-mlops-portfolio/rule-audit.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 | $0.00039 | $0.02224 |
| Opus 5 | $0.00019 | $0.01112 |
| Sonnet 5 | $0.00008 | $0.00445 |
| Haiku 4.5 | $0.00004 | $0.00222 |
Grade A, and why
rule-audit 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 8d 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 — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rule Audit — Compliance scan against AGENTS.md invariants
This skill is READ-ONLY. It does not modify code. When findings are identified, the agent produces a remediation plan and waits for human approval (CONSULT) before any fix. Template-scaffolder patches require STOP because they affect all future services.
When NOT to use this skill
- Before running tests — broken tests are a faster signal than structural audits.
- During active incidents — use
/incidentor/rollbackfirst. Come back to audit in the post-mortem. - On third-party code — this skill is scoped to the template's own conventions.
Invariant catalogue (D-01 → D-27)
The checks below map one-to-one onto the AGENTS.md anti-pattern table. The agent runs the matching query for each ID and records PASS/FAIL + evidence (file:line or metric) per service.
Serving (D-01 → D-05)
| ID | Check | Command |
|---|---|---|
| D-01 | uvicorn --workers N where N > 1 |
rg -n "uvicorn.*--workers\s+[2-9]" --glob "**/Dockerfile*" --glob "**/*.sh" |
| D-02 | HPA uses memory metric | yq '.spec.metrics[] | select(.resource.name == "memory")' k8s/**/hpa.yaml |
| D-03 | model.predict( inside an async def without run_in_executor |
rg -n "async def.*:" -A 30 app/ | rg "model\.(predict|predict_proba)\(" |
| D-04 | shap.TreeExplainer used with stacking/ensemble wrapper |
rg -n "TreeExplainer" app/ |
| D-05 | == in requirements.txt for ML packages |
rg -n "^(numpy|scipy|scikit-learn|pandas|xgboost|lightgbm)==" requirements*.txt |
Infrastructure + Quality (D-06 → D-12)
| ID | Check |
|---|---|
| D-06 | Any metric in artifacts/metrics.json ≥ 0.99 — requires ADR |
| D-07 | Fairness DIR < 0.80 or missing |
| D-08 | PSI CronJob reports but no alert routes configured |
| D-09 | Drift CronJob last_run > 48h → heartbeat missing |
| D-10 | terraform.tfstate* tracked in git |
| D-11 | Model artifact present in Dockerfile (COPY *.joblib / *.pkl) |
| D-12 | Promotion pipeline missing a quality-gate script |
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.
- 8d ago First seen · 209 lines · 39 tokens per session scan A 08bcd66431a5
rule-audit is a skill published in the GitHub repository DuqueOM/ML-MLOps-Portfolio (5 stars, last pushed yesterday), licensed MIT. It adds 39 tokens to every session and 2,224 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
Cloud Security & Container Hardening
AWS/Azure/GCP security auditing, container and Kubernetes hardening, Infrastructure as Code scanning, and cloud compliance assessment.
status
Show the current MLflow tracing configuration for Claude Code.
mlflow
Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow - framework-agnostic ML lifecycle platform.
setup
Configure MLflow tracing for Claude Code.
ecspresso
ECS deployment tool - deploy, manage, and troubleshoot ECS services.
implementing-container-image-minimal-base-with-distroless
Reduce container attack surface by building application images on Google distroless base images that contain only the application runtime with no shell, package manager, or unnecessary OS utilities.