Merge a pull request.
AI agents use merge_pr to create or update resources in Downscoping — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Downscoping environment.
Merging a pull request creates/modifies repository state (commits are added to the target branch). While theoretically reversible via revert commits, the blast radius is high in a CI/CD context—malicious or buggy code merged to main could compromise production systems, affect all downstream consumers, or trigger automated deployments.
From the tool's definition Tool named 'merge_pr' with description 'Merge a pull request.' Merging a PR modifies the repository state by integrating code changes into a branch, typically main/master.
Attacks that exploit this kind of access
Merge a pull request. It is categorised as a Write tool in the Downscoping MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Downscoping MCP server in PolicyLayer and add a rule for merge_pr: 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 Downscoping. Nothing to install.
merge_pr 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 merge_pr 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 merge_pr. 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.
merge_pr is provided by the Downscoping MCP server (kbroughton/downscoping-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 →