search_prompts
AI agents call search_prompts to retrieve information from Prompt Store MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Search operations retrieve and filter existing data without modification, deletion, or execution of external operations. The tool operates on a local-only SQLite database of prompts, limiting blast radius to information disclosure of the user's own prompt library.
From the tool's definition Tool name 'search_prompts' indicates a query/retrieval operation on a local SQLite prompt library. The server description emphasizes 'store, search, and manage' prompts, with search being a read operation.
Attacks that exploit this kind of access
search_prompts. It is categorised as a Read tool in the Prompt Store MCP MCP Server, which means it retrieves data without modifying state.
Register the Prompt Store MCP server in PolicyLayer and add a rule for search_prompts: 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 Prompt Store MCP. Nothing to install.
search_prompts 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_prompts 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_prompts. 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_prompts is provided by the Prompt Store MCP server (mhoshdev/prompt-store-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 →