Menu
Scan the current repo and score its feedback loop maturity for AI-assisted development
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Add a new linter to vigiles's cross-referencing engine as one cohesive, type-enforced unit — a LinterAdapter in the LINTERS registry, with the conformance test enforcing docs + site parity so no site is forgotten
Deep linter reference for authoring or debugging a vigiles enforce() rule — plugin tables, AST selectors, type-aware rules, auto-fix, and edge cases for ESLint, Ruff, Pylint, RuboCop, Stylelint, and Clippy. Use when you need the exact rule name or config for a specific linter, not for running a linter. (JVM/Go linters — detekt, ktlint, Checkstyle, golangci-lint — and Cedar have no deep-dive file yet; their reference lives in docs/linter-support.md.)
Hunt for real bugs in the vigiles CLI/codebase with a parallel expert fan-out, then FIX them directly — source-trace each defect to file:line, add a regression test, commit per theme. Use when asked to dogfood vigiles, find/fix bugs across the CLI the source-traced way, or fan out agents to audit the tool on itself.
The purpose, UX flow, and modern-minimalist design bar for the vigiles.sh landing site (site/). Use whenever designing, editing, adding to, or reviewing any site/ component, hero, section, or marketing copy — read it BEFORE touching site/, and hold every change against it. Not for docs/ or the app itself.
Render a local web page / built site and capture full-page screenshots (desktop + mobile) using the pre-installed Chromium. Use when asked to screenshot, render, or "see how it looks" for the vigiles landing site (site/), the audit report (report/), or any local HTML/dev server. Handles the scroll-reveal gotcha (below-the-fold cards captured blank) automatically.
Synthesize a recurring code-review rule into a custom lint rule — gated by an independent soundness test that abstains rather than ship a checker it can't prove sound
| name | audit-feedback-loop |
| description | Scan the current repo and score its feedback loop maturity for AI-assisted development |
| disable-model-invocation | true |
Scan the current repository and score its feedback loop maturity for AI-assisted development.
Analyze this repository and score its feedback loop maturity using the levels below. Check for each signal, then output a summary report.
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.
Scan the repository for the following and note which exist:
.github/workflows/, .circleci/, Jenkinsfile, .gitlab-ci.yml, bitbucket-pipelines.yml, .travis.yml, etc.eslint.config.*, .eslintrc*, biome.json, .prettierrc*, deno.jsonpyproject.toml (look for [tool.ruff], [tool.pylint], [tool.flake8]), setup.cfg, .flake8, ruff.tomlclippy.toml, .clippy.toml, rustfmt.toml.golangci.yml, .golangci.yaml.rubocop.ymlcheckstyle.xml, pmd.xml, detekt.ymleslint-plugin-*, eslint-rules/ directoriesCLAUDE.md exists at the repo rootCLAUDE.md.spec.ts exists) or v1 annotations (**Enforced by:** in CLAUDE.md). v2 specs = higher maturity.CLAUDE.md.spec.ts or *.spec.ts files — indicates typed spec compilation via vigiles v2.vigiles/generated.d.ts — indicates linter rules are type-checked at authoring timeplaywright.config.*), Cypress (cypress.config.*), Chromatic, Percy, BackstopJS configs@sentry/, dd-trace, @datadog/, newrelic, @opentelemetry/, sentry_sdk, structlog, tracing (Rust), opentelemetry in source files.github/workflows/ with schedule: triggers, or references to scheduled Claude Code tasks## Feedback Loop Audit
**Repository:** <repo name>
**Primary language(s):** <detected languages>
**Score: Level X — <Name>**
### Signals Found
- [x] CI Configuration: <details>
- [ ] Custom Lint Rules: not found
- [x] CLAUDE.md: found, 5 enforced / 2 guidance / 1 missing
...
### Recommendations
1. <Most impactful next step to level up>
2. <Second recommendation>
3. <Third recommendation>
### How to Level Up
<Specific, actionable advice for reaching the next maturity level>
Be specific about file paths and what you found. Give actionable recommendations tailored to the project's language and toolchain.