Search commits by keyword across commit messages (and optionally code changes). Returns matching commits annotated with their pull requests and linked issues. Useful for finding when a feature was added, a bug was fixed, or a dependency was changed.
AI agents call search_commits to retrieve information from Git Context without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a read-only tool that queries git history and commit messages without side effects. It retrieves and filters existing data from a git repository's commit log. The search operation itself is non-destructive and returns only information. While it could theoretically expose sensitive information in commit messages, the primary function is benign data retrieval, making it a Read risk with low severity.
From the tool's definition Tool searches commits and returns matching results with metadata (pull requests, linked issues). The description uses query language: 'search commits by keyword', 'returns matching commits', 'finding when' — all passive retrieval operations.
Attacks that exploit this kind of access
Search commits by keyword across commit messages (and optionally code changes). Returns matching commits annotated with their pull requests and linked issues. Useful for finding when a feature was added, a bug was fixed, or a dependency was changed. It is categorised as a Read tool in the Git Context MCP Server, which means it retrieves data without modifying state.
Register the Git Context MCP server in PolicyLayer and add a rule for search_commits: 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 Git Context. Nothing to install.
search_commits 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 search_commits 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 search_commits. 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.
search_commits is provided by the Git Context MCP server (muhannad-hash/git-context-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 →