AI agents use github_create_pull_request to create or update resources in Github — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Github environment.
Creating a pull request is a Write action: it creates new repository state (the PR object, associated metadata, triggers CI/CD pipelines, and notifies reviewers) but is fully reversible (PRs can be closed without merging). It does not execute code, delete data irreversibly, or move money.
From the tool's definition Tool creates a pull request, which is a persistent, reversible modification to repository metadata. Description states 'Create a ready-for-review pull request from a pushed feature branch into the base branch.'
Attacks that exploit this kind of access
Create a ready-for-review pull request from a pushed feature branch into the base branch. It is categorised as a Write tool in the Github MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Github MCP server in PolicyLayer and add a rule for github_create_pull_request: 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 Github. Nothing to install.
github_create_pull_request 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 github_create_pull_request 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 github_create_pull_request. 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.
github_create_pull_request is provided by the Github MCP server (matiasnjacob/github-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 →