vc-audit-context

Validates context routing, skill discoverability, and Claude/Codex wiring with Node.js validator scripts.

2|Updated Sep 15, 2025
One-click install
npx skills add https://github.com/marsley01/Marsley-Portfolio-Web --skill vc-audit-context-marsley01
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vc-audit-context
Source: https://github.com/marsley01/Marsley-Portfolio-Web/tree/main/.agents/skills/vc-audit-context
Command: npx skills add https://github.com/marsley01/Marsley-Portfolio-Web --skill vc-audit-context-marsley01

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Project context documentation and shared skills drift over time: routers stop indexing files, skills become unreachable from discovery surfaces, and Claude/Codex agent definitions fall out of sync. This Skill audits the durable context layer and skill inventory to detect broken references, unindexed docs, and routing gaps before they cause silent failures. ## Core Features & Use Cases - Context Discovery Validation: Runs validators that confirm every process/context/ doc is indexed by the root router or its group entrypoint, and that concrete backticked references resolve to real files. - Skill Routing & Catalog Checks: Verifies every shared skill is routed from canonical surfaces or explicitly allowlisted, enforces trigger_keywords and layer frontmatter, and keeps the generated skills catalog in sync. - Cross-Surface Parity: Checks Claude and Codex agent name parity, skill cross-references, dependency cycles, and confusable skill names. - Use Case: After reorganizing your project's process/context/ docs or adding a new shared skill, run this audit to confirm the router, group entrypoints, and .agents/skills symlink all still resolve correctly. ## Quick Start Ask the AI to audit the project context routing and skill discoverability, then fix any validator failures it reports.

Frequently Asked Questions about vc-audit-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit context routing in a Claude/Codex project?

Run the validator scripts under .claude/skills/vc-audit-context/scripts/, starting with validate-context-discovery.mjs. It checks that every process/context doc is indexed by the root router or its group entrypoint and that concrete references resolve.

How do I check that shared skills are discoverable by Codex?

The audit confirms .agents/skills resolves to .claude/skills and that every skill folder has a SKILL.md reachable through both paths. The validate-skill-routing.mjs script also ensures each skill is routed from canonical surfaces or explicitly allowlisted.

What does the generated skills catalog do?

generate-skills-catalog.mjs writes process/context/generated-skills-catalog.json, a deterministic machine-readable inventory of all skills with aliases, trigger keywords, layers, and routing sources. The --check flag fails if the catalog is stale relative to the current skill inventory.

Why does the context discovery validator skip checks on a fresh clone?

In bare-kit mode, when process/context/all-context.md does not exist yet, per-project context-doc checks are intentionally skipped because the kit template has not been set up. Structural checks on .claude, .codex, and AGENTS.md still run.

What frontmatter is required on every SKILL.md?

Each SKILL.md must carry a name, description, non-empty trigger_keywords, and a layer value of actor, contract, or helper. The validate-skill-keywords.mjs script enforces this and also verifies the generated catalog is in sync.

When should I use audit-vc instead of this audit?

Use the audit-vc skill for agent and skill harness validation such as agent parity, skill frontmatter schema, README.md sync, and protocol wiring. This skill focuses on context routing, skill discoverability, and catalog consistency.