AI agents call ism_stats to retrieve information from Ism without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and reports database state information without creating, modifying, executing, or deleting data. It is a passive read-only query operation with minimal blast radius—miscall or abuse could only expose metadata about the database structure, not sensitive controls or data.
From the tool's definition Tool name 'ism_stats' and description 'Report database state' indicate a query/reporting function that retrieves metadata about the ISM database (active version, control counts, db path) with no modification or execution capability.
Attacks that exploit this kind of access
Report database state: active ISM version, version and control counts, db path. It is categorised as a Read tool in the Ism MCP Server, which means it retrieves data without modifying state.
Register the Ism MCP server in PolicyLayer and add a rule for ism_stats: 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 Ism. Nothing to install.
ism_stats 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 ism_stats 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 ism_stats. 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.
ism_stats is provided by the Ism MCP server (samueldudley/ism-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 →