Commit the current draft of a grader, creating a new read-only version. IMPORTANT: Only commit AFTER a successful test_evaluator run. After committing, use create_evaluation_pipeline to wrap the grader in a V2 pipeline that renders in the UI.
AI agents use commit_evaluator to create or update resources in Respan MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Respan MCP Server environment.
This is a Write action because it creates a new version artifact and persists changes to the evaluator/grader. While it transitions a draft to read-only (preventing further modifications to that version), the act of committing itself is a write operation that creates new data. It is not Destructive because the draft version is not deleted—commits are typically additive.
From the tool's definition The tool creates 'a new read-only version' of a grader, which is a persistent creation/modification operation. The description states it 'commit[s]' a draft, which is a version control write operation that transforms mutable state into immutable state.
Attacks that exploit this kind of access
Commit the current draft of a grader, creating a new read-only version. IMPORTANT: Only commit AFTER a successful test_evaluator run. After committing, use create_evaluation_pipeline to wrap the grader in a V2 pipeline that renders in the UI. It is categorised as a Write tool in the Respan MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Respan MCP Server MCP server in PolicyLayer and add a rule for commit_evaluator: 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 Respan MCP Server. Nothing to install.
commit_evaluator 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 commit_evaluator 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 commit_evaluator. 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.
commit_evaluator is provided by the Respan MCP Server MCP server (respanai/respan-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 →