Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Audit VC (Version Control Harness Health) — measured AI agent skill | DecimalAI

withkynam/vc-audit-vc

SkillScore82Maintained

Audit agent harness health: Claude/Codex agent parity, skill registry consistency, README.md sync, and protocol file wiring. Use when agents, skills, README.md, or development-protocol files move, split, or drift.

agent-toolingby withkynam

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

DevOps & SRE pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/withkynam-vc-audit-vc/SKILL.md

decimalai skills pull withkynam-vc-audit-vc --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/withkynam-vc-audit-vc/SKILL.md https://app.decimal.ai/s/withkynam-vc-audit-vc/SKILL.md

Raw file: /s/withkynam-vc-audit-vc/SKILL.md ↗ · machine-readable summary at /s/withkynam-vc-audit-vc.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 3h ago

—-31% tokens0% turns82% of whole cases pass with skill (18/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-04	✗→✓	▲ Improved	-43%	0%
case-05	✗→✓	▲ Improved	-29%	0%
case-07	✗→✓	▲ Improved	-33%	0%
case-09	✗→✓	▲ Improved	-61%	0%
case-10	✗→✓	▲ Improved	-26%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

withkynam / vibecode-pro-max-kit

.claude/skills/vc-audit-vc/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1Bundled 9 files · scripts

Skill contents

scripts/

SKILL.md

Audit VC (Version Control Harness Health)

> Output style: Follow process/development-protocols/communication-standards.md — answer-first, plain language, no unexplained jargon, TL;DR on long responses.

Use this skill to verify that the agent harness layer is internally consistent and correctly wired across Claude, Codex, README.md, and protocol files.

For context routing, grouping, and discoverability audits, use the audit-context skill instead.

Workflow

Run the Claude/Codex agent parity validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-agent-parity.mjs

Run the shared skill discovery validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-skills.mjs

Run the README.md sync validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-guide-sync.mjs

Run the protocol wiring validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-protocol-wiring.mjs

Run the seed/scaffold consistency validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-seeds.mjs

Run the kit portability validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-kit-portability.mjs

Run the skill invocation wiring validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-skill-invocation-wiring.mjs

Run the agent frontmatter validator:

bash node .claude/skills/vc-audit-vc/scripts/validate-agent-frontmatter.mjs

If any script reports failures, inspect the referenced files and patch the smallest

relevant surface.

Re-run the failed validators until they pass.

Rules

Treat .claude/agents/ as canonical for agent definitions; .codex/agents/ mirrors them.
Treat .claude/skills/ as canonical for skills; .agents/skills/ is the Codex discovery symlink.
When updating agents, mirror Claude markdown and Codex TOML surfaces together.
Treat process/_seeds/ as an optional legacy scaffold surface in the live repo. Its absence is a warning-only audit result unless the user is explicitly auditing export-kit scaffolding.
Treat validator warnings as audit findings unless the user asks for a strict cleanup.
For context routing and discoverability audits, delegate to audit-context.

Want to see if withkynam/vc-audit-vc would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.