eval-audit

A command for checking every SKILL.md file under a directory. SKILL.md files contain the instructions that coding-agent skills follow.

In plain words
What is it for?
Finding skills recursively, running full or security-only evaluations, producing JSON or Markdown reports, and optionally fixing failures within a token budget.
Why use it?
It replaces one-by-one review with a batch report and can identify security or quality failures across a whole collection of skills.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/evol-ai/skillcompass/eval-audit
Clone the repo
git clone --depth 1 https://github.com/Evol-ai/SkillCompass
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,604 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01604
Opus 5 $0.00000 $0.00802
Sonnet 5 $0.00000 $0.00321
Haiku 4.5 $0.00000 $0.00160

Measured 2d ago against content hash 3aed92b257bd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

eval-audit scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Enumerates other installed skillslowAgent snooping

Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.

Use the **Glob** tool to find all `**/SKILL.md` files recursively under the specified directory. Also check `~/.claude/skills/` if scanning project-level.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

commands/eval-audit.md · 155 lines

How it starts

The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/eval-audit — Batch Skill Evaluation

Locale: All templates in this spec are written in English. Detect the user's language from the session and translate user-facing text at display time per SKILL.md's Global UX Rules. Dimension labels: see the canonical table in SKILL.md.

Arguments

  • <directory> (optional, default: .claude/skills/): Directory to scan for skills.
  • --security-only (optional): Only run D3 security scan per skill.
  • --format [json|md|all] (optional, default: json): Output format.
  • --fix (optional): After evaluation, improve each FAIL skill. Requires --budget.
  • --budget <number> (required with --fix): Maximum total token estimate for improvements. Example: --budget 200000.
  • --fix-caution (optional): Also improve CAUTION skills (only with --fix).
  • --ci (optional): CI-friendly mode. Suppresses interactive prompts, outputs JSON only, sets exit code (0=all PASS, 1=any CAUTION, 2=any FAIL).
  • --internal (optional): Called by another command. Skip all interactive prompts and return results only.

Steps

Step 1: Discover Skills

Use the Glob tool to find all **/SKILL.md files recursively under the specified directory. Also check ~/.claude/skills/ if scanning project-level.

Exclude: test-fixtures/, node_modules/, archive/, .git/, .skill-compass/.

If no SKILL.md files found: display the message, then — unless --internal or --ci is set — offer:

No skills found in {directory}.
  [Run /setup to check global] [Specify another directory] [Cancel]

If --internal or --ci: output the error message only (no choices) and stop.

Step 2: Evaluate Each Skill

For each discovered SKILL.md, display progress — unless --internal is set, in which case output minimal machine-readable progress only:

[{N}/{total}] Evaluating: {skill-name}...
  • Full mode (default): use the Read tool to load {baseDir}/commands/eval-skill.md and follow its evaluation flow for each skill, passing --internal.
  • Security-only mode (--security-only): use the Read tool to load {baseDir}/commands/eval-security.md and follow its flow for each skill, passing --internal.

Read the full file on GitHub · 155 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 155 lines · 0 tokens per session scan A 3aed92b257bd

Subscribe to this mod's changes

eval-audit is a command published in the GitHub repository Evol-ai/SkillCompass (215 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,604 tokens. A static security scan graded it A with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.