🎯

audit-feedback-loop

🎯Skill

from zernie/vigiles

Installation

Vibe Index InstallInstalls to .claude/skills/

$npx vibeindex add zernie/vigiles --skill audit-feedback-loop

skills.sh Install⚠ Installs to .agents/skills/

$npx skills add zernie/vigiles --skill audit-feedback-loop

Manual InstallCopy SKILL.md content and save to the path below

path

~/.claude/skills/audit-feedback-loop/SKILL.md

What it does

A verification tool for CLAUDE.md and instruction files that checks whether referenced linter rules, file paths, scripts, and code symbols actually exist and are current, preventing agents from acting on stale documentation.

📦

Same repository

zernie/vigiles(18 items)

audit-feedback-loop

SKILL.md

313Installs

AddedJun 9, 2026

View on GitHub Back to Skills

More from this repository10

🎯

edit-spec🎯Skill

A tool that verifies references in CLAUDE.md and other instruction files against real project state, catching stale file paths, disabled linter rules, renamed scripts, and missing code symbols. Offers three adoption levels from inline comments to a full TypeScript config with CI integration.

🎯

strengthen🎯Skill

A vigiles skill that upgrades guidance() rules to compile-verified enforce() rules by reading linter-specific reference docs, helping convert prose coding conventions into mechanically enforced lint rules across ESLint, Stylelint, Ruff, Clippy, and more.

Skill

Skill

Skill

Skill

Skill

debug-my-harness🎯Skill

Skill

🎯

pr-to-lint-rule🎯Skill

Part of vigiles, a tool that detects when CLAUDE.md instructions drift from actual project configuration by verifying documented rules against real linter configs, scripts, and file paths to prevent agents from following outdated guidance.

🎯

enforce-rules-format🎯Skill

Part of vigiles, a tool that verifies references in your AI agent instruction files (CLAUDE.md). This skill validates that all rules have proper enforcement classification, ensuring linter rules, file paths, and code symbols referenced in your instructions actually exist.