AI agents call pr_changes to retrieve information from Pr Review without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves data about pull request comments or changes without modifying any state. It follows a read pattern typical of pagination/incremental fetch operations. No creation, modification, deletion, or external execution is implied. The low severity reflects minimal risk—the worst case is exposure of PR comment data already visible within the repository context.
From the tool's definition Tool name 'pr_changes' and description 'Get comments since cursor for incremental updates' indicate a retrieval operation. The verb 'Get' and the concept of fetching updates since a cursor position are characteristic of read-only queries with no side effects.
Attacks that exploit this kind of access
Get comments since cursor for incremental updates. It is categorised as a Read tool in the Pr Review MCP Server, which means it retrieves data without modifying state.
Register the Pr Review MCP server in PolicyLayer and add a rule for pr_changes: 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 Pr Review. Nothing to install.
pr_changes is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the pr_changes 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 pr_changes. 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.
pr_changes is provided by the Pr Review MCP server (thebtf/pr-review-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 →