AI agents call github_pull_main to retrieve information from Github without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs git fetch and pull operations, which retrieve updates from a remote repository but do not modify, delete, or create any data in the repository itself. It is a standard developer workflow operation for synchronizing local branches with upstream changes. The fast-forward pull constraint ensures no destructive merges occur.
From the tool's definition Tool description explicitly states 'Fetch, switch to the base branch, and fast-forward pull from the configured remote.' These are read and navigation operations with no data modification, deletion, or code execution beyond standard git operations.
Attacks that exploit this kind of access
Fetch, switch to the base branch, and fast-forward pull from the configured remote. It is categorised as a Read tool in the Github MCP Server, which means it retrieves data without modifying state.
Register the Github MCP server in PolicyLayer and add a rule for github_pull_main: 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_pull_main 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 github_pull_main 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_pull_main. 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_pull_main 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 →