Scan AI agent config files (CLAUDE.md, AGENTS.md, .cursorrules, etc.) for stale references, dead paths, token bloat, and (when include_drift is set) drift between agent rules and the live MCP tool / skill / command surface. Read-only. Returns JSON: { issues: [{ file, line, category, issue, severi...
AI agents call audit_config to retrieve information from Trace without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a read-only scan of configuration files and returns analysis results. It does not modify, delete, or execute anything. Severity is medium because it reads potentially sensitive AI agent configuration files (CLAUDE.md, AGENTS.md, .cursorrules) which could reveal system prompts, tool surfaces, and internal agent rules if misused.
From the tool's definition 'Read-only. Returns JSON: { issues: [...], total }' and 'Scan AI agent config files ... for stale references, dead paths, token bloat, and drift'
Documented attack patterns abuse exactly the kind of access audit_config gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Trace, and nothing reaches the server without passing your rules. This is the rule we recommend for audit_config:
{
"version": "1",
"default": "deny",
"tools": {
"audit_config": {}
}
}audit_config is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Scan AI agent config files (CLAUDE.md, AGENTS.md, .cursorrules, etc.) for stale references, dead paths, token bloat, and (when include_drift is set) drift between agent rules and the live MCP tool / skill / command surface. Read-only. Returns JSON: { issues: [{ file, line, category, issue, severity, fix? }], total }. It is categorised as a Read tool in the Trace MCP Server, which means it retrieves data without modifying state.
Register the Trace MCP server in PolicyLayer and add a rule for audit_config: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Trace. Nothing to install.
audit_config is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the audit_config rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for audit_config. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
audit_config is provided by the Trace MCP server (nikolai-vysotskyi/trace-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 178 Trace tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
178 Trace tools catalogued and risk-classified — across an index of 42,500+ MCP servers.