memory_set

THE RISK

Medium Risk

Why memory_set needs a policy

This tool creates or modifies data in a shared memory store used by multiple agents. While reversible (values can be overwritten), it affects shared state that other agents depend on for coordination. Misuse could corrupt shared context, inject false decisions, or block legitimate task progression. This is Write rather than Execute because it does not run code or trigger external operations—it only persists data.

From the tool's definition Tool description states 'Store a value in shared memory', which is a create/modify operation. The tool accepts arguments to write data into a shared namespace system (context, decisions, findings, blockers).

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

POLICY

How to control memory_set

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

policy.json

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

memory_set 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 Agent Orchestration — 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 Agent Orchestration tools

Destructive memory_delete Delete a value from shared memory. Destructive agent_unregister Unregister this agent and release all held locks. Call at the end of your session. Execute bootstrap Initialize agent session: register (if needed), get current focus, pending tasks, and rece Execute cursor_recover_task Recover a failed or stale delegated Cursor task by relaunching it with the latest shared c Execute lock_acquire Acquire a lock on a resource to prevent concurrent access. Write claim_todo FOR SUB-AGENTS: Register yourself AND claim a specific task in one call. Use this when you Write cursor_delegate_task Delegate a task to Cursor CLI and store session metadata for later resume/sync. Write cursor_sync_task Refresh a delegated Cursor task, sync the latest provider status, and persist knowledge ba

All 35 Agent Orchestration 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 memory_set

What does the memory_set tool do? +

Store a value in shared memory. Use namespaces to organize: context, decisions, findings, blockers. It is categorised as a Write tool in the Agent Orchestration 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_set? +

Register the Agent Orchestration MCP server in PolicyLayer and add a rule for memory_set: 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 Agent Orchestration. Nothing to install.

What risk level is memory_set? +

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

Can I rate-limit memory_set? +

Yes. Add a rate_limit block to the memory_set 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_set completely? +

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

memory_set is provided by the Agent Orchestration MCP server (madebyaris/agent-orchestration). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Agent Orchestration tool call.

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

CHECK YOUR STACK →