AI agents call search to retrieve information from Recap 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 from the workspace without creating, modifying, deleting, or executing any operations. It has no side effects and presents minimal risk—the only concern would be if sensitive data is indexed, but that is a data sensitivity issue, not a tool risk category issue. Classified as Read with low severity and high confidence.
From the tool's definition Tool description states 'Fuzzy ranked search across projects, tasks, sessions, and notes' with explicit purpose to 'recall anything in the workspace.' The verb 'search' and function of querying/retrieving existing data with no modification capability indicate…
Attacks that exploit this kind of access
Fuzzy ranked search across projects, tasks, sessions, and notes. Ranks by recency, type relevance, and field match. Use to recall anything in the workspace. It is categorised as a Read tool in the Recap MCP Server, which means it retrieves data without modifying state.
Register the Recap MCP server in PolicyLayer and add a rule for search: 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 Recap. Nothing to install.
search 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 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. 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 is provided by the Recap MCP server (shivam-singh-git/recap-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 →