search_chats
AI agents call search_chats to retrieve information from Cursor Chats Bridge without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Despite empty tool description, the server's explicit read-only nature and the pattern of sibling tools that query/list data without side effects clearly indicate search_chats is a retrieval operation. No data modification, deletion, execution, or financial operations are possible on this bridge. Low severity because misuse would only retrieve existing chat data without further impact.
From the tool's definition Tool name 'search_chats' combined with server description stating 'Read-only bridge' and 'Exposes Cursor's local chat database via 7 MCP tools'.
Attacks that exploit this kind of access
search_chats. It is categorised as a Read tool in the Cursor Chats Bridge MCP Server, which means it retrieves data without modifying state.
Register the Cursor Chats Bridge MCP server in PolicyLayer and add a rule for search_chats: 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 Cursor Chats Bridge. Nothing to install.
search_chats 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_chats 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_chats. 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_chats is provided by the Cursor Chats Bridge MCP server (jherard-fr/cursor-chats-bridge). 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 →