AI agents use project_undo to create or update resources in ReaperMCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your ReaperMCP environment.
Undo is categorized as Write rather than Destructive because it restores previous states rather than irreversibly deleting data. However, it modifies project state, so it ranks above Read. In a music production context, an AI agent invoking undo could discard user work unexpectedly, creating moderate blast radius (lost edits), justifying medium severity.
From the tool's definition Tool is named 'project_undo' with description 'Undo last action.' It reverses modifications to project state by restoring a prior state, which is a write-class operation that modifies data reversibly.
Documented attack patterns abuse exactly the kind of access project_undo gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and ReaperMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for project_undo:
{
"version": "1",
"default": "deny",
"tools": {
"project_undo": {
"limits": [
{
"counter": "project_undo_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}project_undo 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.
Free to start. No card required.
Undo last action. It is categorised as a Write tool in the ReaperMCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Reaper MCP server in PolicyLayer and add a rule for project_undo: 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 ReaperMCP. Nothing to install.
project_undo is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the project_undo 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.
Set action: deny in the PolicyLayer policy for project_undo. 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.
project_undo is provided by the Reaper MCP server (xdarkzx/reaper-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from ReaperMCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
138 ReaperMCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.