AgentAudit
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| discover_serversA | Scan local config files to list ALREADY INSTALLED MCP servers (Claude Desktop, Cursor, Windsurf, VS Code). Use ONLY when the user wants to review/list their existing servers. Do NOT use this when the user wants to install, evaluate, or look up a specific package — use check_package for that instead. |
| audit_packageA | Deep security audit of a Git repository. Clones the repo and returns source code with a 3-pass audit methodology (UNDERSTAND → DETECT → CLASSIFY). You then analyze the code and call submit_report with findings. Use check_package FIRST to see if an audit already exists — only use this for unaudited packages or when a fresh audit is requested. |
| submit_reportA | Submit a completed security audit report to the AgentAudit registry (agentaudit.dev). Call this after you have analyzed the code from audit_package. The report becomes publicly available and helps other agents make install decisions. |
| check_packageA | Look up a package in the AgentAudit security registry. USE THIS FIRST whenever the user wants to install, add, evaluate, or learn about a specific MCP server or package. Returns risk score, findings, and official audit status if available. If the package is not yet in the registry, suggests running an audit. This is the go-to tool for any "is this safe?" or "should I install this?" question. |
| consensus_analysisA | Get multi-auditor consensus analysis for a package. Shows how many different auditors and models have scanned the package, their agreement on findings, and the overall consensus risk level. Useful for getting a second opinion on any package's security posture. |
| search_packagesA | Search the AgentAudit security registry for packages. Returns matching packages with their risk scores and audit status. Use this when you need to find a package but don't know the exact name. |
| scan_tool_poisoningA | Scan MCP tool definitions for hidden instructions, unicode tricks, obfuscated payloads, and manipulation patterns. Use this to check if a server's tools contain poisoning indicators (prompt injection in descriptions, zero-width characters, cross-tool manipulation, homoglyph attacks). Provide tool definitions directly OR a source_url to extract them from code. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Most tools have clearly distinct purposes: discover_servers is for listing installed servers, audit_package for deep code audits, check_package for registry lookups, etc. However, check_package and consensus_analysis both relate to registry lookups and could be confused, though descriptions clarify that one is for initial checks and the other for multi-auditor consensus.
Six of the seven tools follow a consistent verb_noun pattern (discover_servers, audit_package, submit_report, check_package, search_packages, scan_tool_poisoning). The one deviation is 'consensus_analysis', which uses a noun phrase instead of a verb_noun form, creating a minor inconsistency.
With 7 tools, the set is well-scoped for the security-audit domain. Each tool serves a distinct stage (discovery, lookup, search, audit, submission, consensus, poisoning scan), fitting comfortably within the ideal 3-15 range without bloat or deficiency.
The toolset covers the full workflow from discovering installed servers to auditing, submitting reports, and obtaining consensus. While there is no tool to edit or delete submitted reports, this is a minor gap and the core lifecycle is complete.