Undo last action
AI agents use undo to create or update resources in Filopastry — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Filopastry environment.
The undo function reverts prior changes to music data or state, which is a Write category operation since it modifies data (by reverting it) in a reversible manner. It does not permanently destroy data (Destructive), execute external commands (Execute), or involve financial transactions. The severity is low because undo is a standard, safe interaction pattern that corrects mistakes rather than introducing new risk.
From the tool's definition Tool name is 'undo' with description 'Undo last action'. This reverses a prior modification to the music pattern or composition state, making it a reversible write operation.
Attacks that exploit this kind of access
Undo last action. It is categorised as a Write tool in the Filopastry MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Filopastry MCP server in PolicyLayer and add a rule for 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 Filopastry. Nothing to install.
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 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 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.
undo is provided by the Filopastry MCP server (youwenshao/filopastry). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →