compass_audit
AI agents call compass_audit to retrieve information from Tool Compass without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Without a description, classification relies on the 'audit' convention which suggests information retrieval and review of logs or access records—a Read operation. The severity is medium because audit tools often access sensitive records (logs, user actions, system events) which could expose operational details if misused by an agent.
From the tool's definition Tool name is 'compass_audit' with empty description. Based on naming convention within the Tool Compass server context, audit tools typically retrieve or query audit logs, access records, and status information rather than modifying data.
Attacks that exploit this kind of access
compass_audit. It is categorised as a Read tool in the Tool Compass MCP Server, which means it retrieves data without modifying state.
Register the Tool Compass MCP server in PolicyLayer and add a rule for compass_audit: 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 Tool Compass. Nothing to install.
compass_audit 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 compass_audit 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 compass_audit. 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.
compass_audit is provided by the Tool Compass MCP server (mcp-tool-shop-org/tool-compass). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →