Filter events by session_id, event_type, agent_id, or date range. Returns paginated events. input_hash is never returned.
AI agents call query_events to retrieve information from Audit event mcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
query_events retrieves and filters audit log data with no side effects. It queries existing records by session_id, event_type, agent_id, or date range and returns results. No data is created, modified, deleted, or executed. This is a straightforward read operation on an audit log.
From the tool's definition Tool description states 'Filter events' and 'Returns paginated events' with no mention of modification, deletion, or execution. The explicit note that 'input_hash is never returned' further confirms read-only access with built-in redaction.
Attacks that exploit this kind of access
Filter events by session_id, event_type, agent_id, or date range. Returns paginated events. input_hash is never returned. It is categorised as a Read tool in the Audit event mcp MCP Server, which means it retrieves data without modifying state.
Register the Audit event MCP server in PolicyLayer and add a rule for query_events: 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 Audit event mcp. Nothing to install.
query_events 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 query_events 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 query_events. 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.
query_events is provided by the Audit event MCP server (mightbesaad/audit-event-mcp). 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 →