Skip to main content
Glama

Audit an agent skill or AGENTS.md

audit_agent_skill
Read-only

Check an agent skill (SKILL.md with its scripts and references) or AGENTS.md, CLAUDE.md, .cursorrules before loading it: prompt injection, curl | bash, encoded commands, credential theft, exfiltration, hardcoded keys, persistence. For MCP servers use audit_mcp_server. Give files or one url, not both. Returns level, findings (type, severity, file, line, snippet) and per-file fingerprints to send back as pinned. Costs US$0.005 in USDC on Base or Solana (x402); nothing is executed.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "audit_agent_skill"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlNoInstead of files: public URL of one raw text file, e.g. a SKILL.md on raw.githubusercontent.com (github.com/.../blob/... links are converted)
filesNoThe skill files with their relative paths, e.g. [{path: 'SKILL.md', content: '...'}, {path: 'scripts/setup.sh', content: '...'}] (max 50 files, 256 KB in total). Omit when using url
pinnedNoOptional map of file path to the fingerprint returned by a previous audit; changed files are reported

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • changedInput schema / properties / files / description
      Previous value: -"The skill files, e.g. [{path: 'SKILL.md', content: '...'}] (max 50, 256 KB)"New value: +"The skill files with their relative paths, e.g. [{path: 'SKILL.md', content: '...'}, {path: 'scripts/setup.sh', content: '...'}] (max 50 files, 256 KB in total). Omit when using url"
    • changedInput schema / properties / pinned / description
      Previous value: -"Path to fingerprint from a previous audit"New value: +"Optional map of file path to the fingerprint returned by a previous audit; changed files are reported"
    • changedInput schema / properties / url / description
      Previous value: -"Or a public URL of one raw text file"New value: +"Instead of files: public URL of one raw text file, e.g. a SKILL.md on raw.githubusercontent.com (github.com/.../blob/... links are converted)"
  2. Added

TDQS

A4.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations by disclosing that nothing is executed, that the call costs US$0.005 in USDC on Base or Solana via x402, and what the response contains. The only gap is that the annotation set already covers read-only safety, and the payment mechanics could be slightly clearer about when payment is charged.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose, then constraints, then outputs and cost, with no filler sentences. The threat enumeration is dense but each item is functional; minor cost is that the opening sentence is a long clause chain.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description correctly steps in to describe the return shape (level, findings with type/severity/file/line/snippet, fingerprints) and adds cost and safety context. Complete enough to call correctly, though it does not mention limits like the 50-file/256 KB caps that only appear in the schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds real meaning: the files/url either-or constraint and the round-trip semantics of pinned (per-file fingerprints to pin a re-audit). That extra context lifts it above the schema alone.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Names a specific verb and resource ('Check an agent skill ... or AGENTS.md, CLAUDE.md, .cursorrules') and enumerates the exact threat classes it detects, so an agent knows precisely what this tool covers. It explicitly distinguishes itself from the sibling audit_mcp_server by scope.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

States the trigger ('before loading it'), routes MCP servers to audit_mcp_server, and gives the mutual-exclusion rule 'Give files or one url, not both.' Both when-to-use and the alternative path are spelled out rather than implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources