checkpoint · Headroom MCP: Risk & Policy

WHAT IT DOES

What checkpoint does on Headroom

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

WHY IT NEEDS A POLICY

Why checkpoint needs a policy

This tool creates and updates checkpoint data (work progress notes, decisions, next steps) in a reversible manner. There are no destructive effects, no code execution, no financial implications, and no external side effects beyond storing internal state.

From the tool's definition Tool creates or modifies a 'survival note' that persists and is re-injected after context compaction. The description explicitly states 'Save YOUR OWN survival note' and 'Re-injected to you after compaction', indicating the tool writes data (the checkpoint…

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

checkpoint Rate-limited

checkpoint 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": {
    "checkpoint": {
      "limits": [
        {
          "counter": "checkpoint_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 Headroom, so you can run agents against it safely.

FAQ

Questions about checkpoint

What does the checkpoint tool do? +

Save YOUR OWN survival note before context compaction: what you are doing, where you are, decisions made (with why), approaches already ruled out, exact next steps, key values. Re-injected to you after compaction. Call when a [headroom] update says context is running low, or before starting work that will not fit. Latest call wins. It is categorised as a Write tool in the Headroom MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on checkpoint? +

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

What risk level is checkpoint? +

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

Can I rate-limit checkpoint? +

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

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

checkpoint is provided by the Headroom MCP server (tyejcoleman/headroom). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.