AI agents call molmcp_find_capability to retrieve information from Molmcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool appears to search or query a codebase index to discover capabilities and symbols—a read-only operation with no side effects. The empty description and reliance on naming conventions and context from sibling tools introduces some uncertainty, but the pattern of read-only discovery operations on this server strongly suggests this is a simple lookup/search tool.
From the tool's definition Tool name 'find_capability' suggests searching or querying indexed code symbols and relationships. Server description indicates it 'expos[es] indexed code symbols, signatures, and relationships' and enables agents to 'find' tools.
Attacks that exploit this kind of access
molmcp_find_capability. It is categorised as a Read tool in the Molmcp MCP Server, which means it retrieves data without modifying state.
Register the Mol MCP server in PolicyLayer and add a rule for molmcp_find_capability: 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 Molmcp. Nothing to install.
molmcp_find_capability 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 molmcp_find_capability 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 molmcp_find_capability. 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.
molmcp_find_capability is provided by the Mol MCP server (molcrafts/molmcp). 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 →