graph-audit

Validate SKILL.md frontmatter, audit-state.json sidecars, and Skill Graph manifest coherence.

Updated May 14, 2026
One-click install
npx skills add https://github.com/jacob-balslev/skills --skill graph-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: graph-audit
Source: https://github.com/jacob-balslev/skills/tree/main/skills/quality-assurance/graph-audit
Command: npx skills add https://github.com/jacob-balslev/skills --skill graph-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Auditing and validating the Skill Graph library to prevent silent drift between SKILL.md frontmatter, audit-state.json sidecars, and the generated manifest. It ensures schema conformance, sidecar integrity, proper manifest joins, and real sibling relation targets across the project.

Core Features & Use Cases

  • Schema conformance checks against the live Skill Metadata Protocol schema files.
  • Sidecar consistency verification for audit-state.json and related loop-owned state.
  • Manifest synchronization to ensure SKILL.md, audit-state.json, and manifest outputs align.
  • Relation integrity checks that relation targets point to real sibling skills.
  • Grounding presence checks for project-targeted skills and name/directory parity across the library.

Quick Start

Run the graph-audit lint to validate SKILL.md, audit-state.json, and manifest coherence across the Skill Graph library.

Frequently Asked Questions about graph-audit

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

FAQPage Schema
How do I audit Skill Graph metadata for schema conformance and manifest drift?

To audit Skill Graph metadata, you validate SKILL.md frontmatter against the live protocol schema, verify audit-state.json sidecar consistency, and check manifest synchronization to report drift across the library.

What causes manifest drift between SKILL.md frontmatter and audit-state.json sidecars?

Manifest drift occurs when SKILL.md frontmatter, audit-state.json sidecars, and the generated manifest lose synchronization. Auditing detects this by enforcing schema validation, sidecar state integrity, and manifest coherence checks.

How do I validate relation integrity to ensure targets point to real sibling skills?

Validating relation integrity involves checking that relation targets defined in the Skill Graph manifest actually point to real sibling skills within the library, ensuring proper manifest joins and preventing broken references.

Can I use graph-audit to check grounding presence and name-directory parity for project-targeted skills?

Yes, when applied to project-targeted skills within the library, the audit enforces grounding presence checks and verifies name-directory parity to ensure all metadata conforms to the Skill Metadata Protocol constraints.

What are the limitations of using schema validation for Skill Graph quality assurance?

Schema validation focuses strictly on enforcing protocol constraints like sidecar state integrity and manifest coherence. It does not fix drift automatically; it only identifies and reports schema conformance issues across the Skill Graph library.