Search memories using semantic similarity. Returns memories most relevant to the query.
AI agents call memory_search to retrieve information from MCP Router without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries stored memories and returns results based on semantic relevance. It has no side effects, does not modify or delete data, and does not execute code or trigger external operations. It is a straightforward search/retrieval function, making it a Read operation.
From the tool's definition Tool name: 'memory_search' with description 'Search memories using semantic similarity. Returns memories most relevant to the query.' — purely a retrieval operation with no modification or deletion of data.
Attacks that exploit this kind of access
Search memories using semantic similarity. Returns memories most relevant to the query. It is categorised as a Read tool in the MCP Router MCP Server, which means it retrieves data without modifying state.
Register the MCP Router MCP server in PolicyLayer and add a rule for memory_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 MCP Router. Nothing to install.
memory_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 memory_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 memory_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.
memory_search is provided by the MCP Router MCP server (justinpreston/mcp-router). 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 →