memory_remember · Tartarus MCP: Risk & Policy

WHAT IT DOES

What memory_remember does on Tartarus

AI agents use memory_remember to create or update resources in Tartarus — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Tartarus environment.

WHY IT NEEDS A POLICY

Why memory_remember needs a policy

This tool writes data to persistent storage (SQLite). It is a Write operation because it creates records that can be undone (deleted) using memory_forget. Misuse could result in storing false, sensitive, or misleading information that persists across sessions and influences future AI behavior, hence medium severity.

From the tool's definition 'Store a memory' — creates a new persistent record in the SQLite database; reversible via the sibling tool memory_forget.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

memory_remember Rate-limited

memory_remember 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.

View as policy code

policy.json

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

LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Tartarus, so you can run agents against it safely.

FAQ

Questions about memory_remember

What does the memory_remember tool do? +

Store a memory. Call after any fact, decision, preference, or context worth keeping across sessions. It is categorised as a Write tool in the Tartarus MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on memory_remember? +

Register the Tartarus MCP server in PolicyLayer and add a rule for memory_remember: 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 Tartarus. Nothing to install.

What risk level is memory_remember? +

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

Can I rate-limit memory_remember? +

Yes. Add a rate_limit block to the memory_remember 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 memory_remember completely? +

Set action: deny in the PolicyLayer policy for memory_remember. 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 memory_remember? +

memory_remember is provided by the Tartarus MCP server (saluca-labs/tartarus-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.