Searches episodic and semantic memory for matching episodes and decisions.
AI agents call cortex_search to retrieve information from Cortex Memory without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries data from the cortex memory system without creating, modifying, deleting, or executing any side effects. It is a pure read operation that returns matching memory records to support context injection for coding sessions.
From the tool's definition Tool name includes 'search' and description states it 'Searches episodic and semantic memory for matching episodes and decisions' — a query operation with no modification or execution of external operations.
Documented attack patterns abuse exactly the kind of access cortex_search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Cortex Memory, and nothing reaches the server without passing your rules. This is the rule we recommend for cortex_search:
{
"version": "1",
"default": "deny",
"tools": {
"cortex_search": {}
}
}cortex_search is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Searches episodic and semantic memory for matching episodes and decisions. It is categorised as a Read tool in the Cortex Memory MCP Server, which means it retrieves data without modifying state.
Register the Cortex Memory MCP server in PolicyLayer and add a rule for cortex_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 Cortex Memory. Nothing to install.
cortex_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 cortex_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 cortex_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.
cortex_search is provided by the Cortex Memory MCP server (skullfire07/cortex-memory). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Cortex Memory, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
5 Cortex Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.