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 skills/zernie/vigiles/audit-feedback-loopnpx skills add zernie/vigiles --skill audit-feedback-loopgit clone --depth 1 https://github.com/zernie/vigilesWrote 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/zernie/vigiles/audit-feedback-loop)<a href="https://agentmods.dev/skills/zernie/vigiles/audit-feedback-loop"><img src="https://agentmods.dev/badge/skills/zernie/vigiles/audit-feedback-loop.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 | $0.00018 | $0.00966 |
| Opus 5 | $0.00009 | $0.00483 |
| Sonnet 5 | $0.00004 | $0.00193 |
| Haiku 4.5 | $0.00002 | $0.00097 |
Grade A, and why
audit-feedback-loop 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 5d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scan the current repository and score its feedback loop maturity for AI-assisted development.
Instructions
Analyze this repository and score its feedback loop maturity using the levels below. Check for each signal, then output a summary report.
Maturity Levels
Level 0 — Vibes No CI config, no linter rules, no CLAUDE.md. The AI agent is flying blind.
Level 1 — Guardrails Has CI + standard linters, but no custom rules. The agent gets basic feedback but can't learn project-specific conventions.
Level 2 — Architecture as Code Has custom lint rules, CLAUDE.md rules have enforcement annotations. The agent gets rich, project-specific feedback.
Level 3 — The Organism Has CI + custom rules + screenshot/visual tests + observability + scheduled agent tasks. The entire development loop is instrumented.
Signals to Check
Scan the repository for the following and note which exist:
- CI Configuration: Look for
.github/workflows/,.circleci/,Jenkinsfile,.gitlab-ci.yml,bitbucket-pipelines.yml,.travis.yml, etc. - Linter Config (language-aware):
- JS/TS:
eslint.config.*,.eslintrc*,biome.json,.prettierrc*,deno.json - Python:
pyproject.toml(look for[tool.ruff],[tool.pylint],[tool.flake8]),setup.cfg,.flake8,ruff.toml - Rust:
clippy.toml,.clippy.toml,rustfmt.toml - Go:
.golangci.yml,.golangci.yaml - Ruby:
.rubocop.yml - Java/Kotlin:
checkstyle.xml,pmd.xml,detekt.yml
- JS/TS:
- Custom Lint Rules: Look for custom plugins, rule directories, or inline rule definitions in linter configs
- JS/TS:
eslint-plugin-*,eslint-rules/directories - Python: custom Ruff/Pylint plugins, AST-based checks
- Rust: custom Clippy lints
- Go: custom analyzers
- JS/TS:
- CLAUDE.md: Check if
CLAUDE.mdexists at the repo root - CLAUDE.md Enforcement: Check if using vigiles v2 specs (
CLAUDE.md.spec.tsexists) or v1 annotations (**Enforced by:**in CLAUDE.md). v2 specs = higher maturity. - Type-Safe Specs: Check for
CLAUDE.md.spec.tsor*.spec.tsfiles — indicates typed spec compilation via vigiles v2 - Generated Types: Check for
.vigiles/generated.d.ts— indicates linter rules are type-checked at authoring time - Screenshot/Visual Tests: Look for Playwright (
playwright.config.*), Cypress (cypress.config.*), Chromatic, Percy, BackstopJS configs - Observability: Search for imports/usage of
@sentry/,dd-trace,@datadog/,newrelic,@opentelemetry/,sentry_sdk,structlog,tracing(Rust),opentelemetryin source files - Scheduled Agent Tasks: Look for cron patterns in CI configs,
.github/workflows/withschedule:triggers, or references to scheduled Claude Code tasks
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.
- 5d ago First seen · 77 lines · 18 tokens per session scan A a9f375d08fea
audit-feedback-loop is a skill published in the GitHub repository zernie/vigiles (15 stars, last pushed yesterday), licensed MIT. It adds 18 tokens to every session and 966 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 2026-08-30.
Other skills, from other repositories
ralplan
Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution.
remember
Review reusable project knowledge and decide what belongs in project memory, notepad, or durable docs.
embedding-strategies
Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.
docx-comment-reply
Reply to comments (批注) in Word .docx/.doc files: extract comment context, draft replies, write threaded replies back, and validate OOXML.
explaining-machine-learning-models
Explain trained machine learning models through feature attribution, local explanations, and behavior summaries. Use as an explicit/manual helper once a model already exists, not for training ownership, leakage auditing, or general ML strategy selection.
nw-ddd-eventsourcing
Event Sourcing and CQRS as DDD implementation patterns — when to use, aggregate event streams, projections, snapshots, sagas, upcasting, conflict resolution.