AI agents call search_clawskills to retrieve information from Cawskills without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries data (skills and playbooks) with ranking logic applied to results. It has no side effects, does not execute code or commands, does not modify data, and does not delete anything. It is a straightforward read operation returning search results. Severity is low because misuse would only return inappropriate query results without affecting system state or data integrity.
From the tool's definition Tool name contains 'search' and description states 'Search across both skills and playbooks' — a query operation with no modification, execution, or deletion of data.
Attacks that exploit this kind of access
Search across both skills and playbooks. For workflow-shaped queries, playbooks are ranked ahead of generic skill matches. It is categorised as a Read tool in the Cawskills MCP Server, which means it retrieves data without modifying state.
Register the Cawskills MCP server in PolicyLayer and add a rule for search_clawskills: 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 Cawskills. Nothing to install.
search_clawskills 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 search_clawskills 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 search_clawskills. 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.
search_clawskills is provided by the Cawskills MCP server (shanksg/clawskills). 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 →