cortex_memory_store

THE RISK

Medium Risk

Why cortex_memory_store needs a policy

This tool writes new memory entries to a persistent store. It creates data that survives across sessions, making it a reversible write operation (memories can presumably be deleted or overwritten). No code execution, financial transactions, or irreversible destruction is implied. Misuse could lead to persistent poisoning of agent memory, hence medium severity.

From the tool's definition 'Store a memory for an AI agent. Memories persist across sessions' — creates/writes persistent data scoped by projectId + branch

Documented attack patterns abuse exactly the kind of access cortex_memory_store gives an agent:

POLICY

How to control cortex_memory_store

PolicyLayer is an MCP gateway — it sits between your AI agents and Cortex Hub, and nothing reaches the server without passing your rules. This is the rule we recommend for cortex_memory_store:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "cortex_memory_store": {
      "limits": [
        {
          "counter": "cortex_memory_store_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

cortex_memory_store stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Cortex Hub — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Cortex Hub tools

Execute cortex_code_reindex Trigger re-indexing of a project after code changes. Looks up the project by repo URL and Execute cortex_cypher Run Cypher queries directly against the GitNexus knowledge graph. Supports MATCH, RETURN, Write cortex_session_start Start a new execution session with optional agent identity metadata. Write cortex_knowledge_store Store a knowledge document in the Cortex knowledge base. Auto-chunks and embeds the conten Write cortex_session_end Close a session with a summary of work done. Reports session duration and compliance. Write cortex_task_accept Accept an assigned task, signaling that work will begin. Updates task status to accepted. Write cortex_task_create Create a task and optionally assign it to another agent. Use for delegating work, tracking Write cortex_task_submit_strategy Submit a task execution strategy for user review. The Lead Agent calls this after analyzin

All 25 Cortex Hub tools →

Write tools on other servers

M-Team MCP Server download_torrent YouTube MCP Server generate_video_title Android Forensics ADB MCP Server combine_reports 0xarchive web3_signup

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Argument validation → Checking tool call arguments against schema and policy before they run.
Scoped token → Per-person credentials that grant a defined subset of servers and tools.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about cortex_memory_store

What does the cortex_memory_store tool do? +

Store a memory for an AI agent. Memories persist across sessions and can be recalled by semantic search. Use projectId + branch to scope memories to a specific branch. It is categorised as a Write tool in the Cortex Hub MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on cortex_memory_store? +

Register the Cortex Hub MCP server in PolicyLayer and add a rule for cortex_memory_store: 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 Hub. Nothing to install.

What risk level is cortex_memory_store? +

cortex_memory_store is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit cortex_memory_store? +

Yes. Add a rate_limit block to the cortex_memory_store 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.

How do I block cortex_memory_store completely? +

Set action: deny in the PolicyLayer policy for cortex_memory_store. 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.

What MCP server provides cortex_memory_store? +

cortex_memory_store is provided by the Cortex Hub MCP server (lktiep/cortex-hub). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Cortex Hub tool call.

Start from Cortex Hub, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →