replay_fork · Maple MCP: Risk & Policy

WHAT IT DOES

What replay_fork does on Maple

AI agents invoke replay_fork to trigger actions in Maple. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why replay_fork needs a policy

This tool executes a replay of agent behavior with modifications, which qualifies as Execute (runs operations whose effects depend on arguments—specifically the edits applied and the step selected). While described as 'dry-run' (suggesting no persistent side effects), the tool still triggers execution logic that could produce different outcomes based on the edits parameter.

From the tool's definition Tool description states 'replay forward in dry-run mode' and 'fork a trace at a selected step, apply edits'.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

replay_fork Rate-limited

replay_fork stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "replay_fork": {
      "limits": [
        {
          "counter": "replay_fork_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

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

FAQ

Questions about replay_fork

What does the replay_fork tool do? +

Fork a trace at a selected step, apply edits, and replay forward in dry-run mode for safe debugging. It is categorised as a Execute tool in the Maple MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on replay_fork? +

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

What risk level is replay_fork? +

replay_fork is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit replay_fork? +

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

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

replay_fork is provided by the Maple MCP server (omar2001ramadan/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.