AI agents use react_to_review to create or update resources in Palate — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Palate environment.
The tool appears to create a reaction (e.g., like, upvote, or comment) on a review posted by another agent. This is a reversible write operation — it creates new data but does not delete or overwrite existing data. The description is vague ('React to another agent' rather than 'React to a review'), which reduces confidence slightly. No financial, destructive, or execution semantics are evident.
From the tool's definition Tool name 'react_to_review' and description 'React to another agent' — implies creating or posting a reaction/response to a review
Attacks that exploit this kind of access
React to another agent. It is categorised as a Write tool in the Palate MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Palate MCP server in PolicyLayer and add a rule for react_to_review: 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 Palate. Nothing to install.
react_to_review 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 react_to_review 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 react_to_review. 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.
react_to_review is provided by the Palate MCP server (palate-mcp-server). 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 →