Prepare commit message context from staged changes. Falls back to unstaged working tree changes when nothing is staged, so you can analyze changes before running
AI agents call generate_commit_message to retrieve information from Pr Narrator without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool reads staged or unstaged git changes to prepare commit message context. It does not actually create a commit, write to the repository, or execute any side-effecting operations. It is an analysis/read operation that surfaces information for the user to act upon.
From the tool's definition 'Prepare commit message context from staged changes' and 'Falls back to unstaged working tree changes when nothing is staged, so you can analyze changes before running' — the tool reads/analyzes git changes to prepare context, with no indication it writes or…
Attacks that exploit this kind of access
Prepare commit message context from staged changes. Falls back to unstaged working tree changes when nothing is staged, so you can analyze changes before running. It is categorised as a Read tool in the Pr Narrator MCP Server, which means it retrieves data without modifying state.
Register the Pr Narrator MCP server in PolicyLayer and add a rule for generate_commit_message: 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 Narrator. Nothing to install.
generate_commit_message 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 generate_commit_message 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 generate_commit_message. 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.
generate_commit_message is provided by the Pr Narrator MCP server (mhaviv/pr-narrator-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 →