AI agents call mcp_find_keys_by_keywords to retrieve information from Memcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool searches or queries stored memory to find matching keys based on keywords. This is a retrieval operation with no side effects - it does not create, modify, delete, or execute anything. The empty description limits confidence slightly, but the naming pattern and server context clearly indicate this is a search/read function typical of data retrieval tools.
From the tool's definition Tool name 'mcp_find_keys_by_keywords' indicates a search/query operation that retrieves keys matching specified keywords.
Attacks that exploit this kind of access
mcp_find_keys_by_keywords. It is categorised as a Read tool in the Memcp MCP Server, which means it retrieves data without modifying state.
Register the Me MCP server in PolicyLayer and add a rule for mcp_find_keys_by_keywords: 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 Memcp. Nothing to install.
mcp_find_keys_by_keywords 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 mcp_find_keys_by_keywords 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 mcp_find_keys_by_keywords. 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.
mcp_find_keys_by_keywords is provided by the Me MCP server (moshez/memcp). 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 →