AI agents use rollback to create or update resources in TDD-MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your TDD-MCP environment.
Rollback modifies the application state (current TDD phase) reversibly—a developer can later re-advance phases if needed. This qualifies as Write rather than Destructive because the change is not irreversible; the previous state can be recovered by advancing again. It is not Execute because it doesn't run arbitrary code or external operations.
From the tool's definition The tool 'rollback' modifies session state by reverting to a previous phase in a TDD cycle, as evidenced by the description 'Rollback to the previous phase' and its presence among session state management tools like 'start_session', 'next_phase',…
Attacks that exploit this kind of access
Rollback to the previous phase. It is categorised as a Write tool in the TDD-MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the TDD- MCP server in PolicyLayer and add a rule for rollback: 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 TDD-MCP. Nothing to install.
rollback 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 rollback 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. 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 is provided by the TDD- MCP server (tinmancoding/tdd-mcp). 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 →