[Interno] Use
AI agents call rollback_partial to permanently remove resources in Maestro — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
A partial rollback operation reverts code or state changes, which can be destructive in nature as it overwrites or removes previously written data. The context of an 'autonomous MCP server for AI-assisted development' with tools like 'atualizar_codebase' and 'auto_fix' suggests this tool modifies the codebase state. Rollbacks in development contexts are generally not easily reversible once applied.
From the tool's definition Tool name 'rollback_partial' implies reverting/undoing changes to a codebase, which is typically an irreversible or hard-to-reverse operation. The description '[Interno] Use' is uninformative.
Attacks that exploit this kind of access
[Interno] Use. It is categorised as a Destructive tool in the Maestro MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Maestro MCP server in PolicyLayer and add a rule for rollback_partial: 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 Maestro. Nothing to install.
rollback_partial is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the rollback_partial 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 rollback_partial. 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.
rollback_partial is provided by the Maestro MCP server (matheus-gama-deluna/maestro). 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 →