AI agents use git_commit to create or update resources in Context — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Context environment.
git_commit creates a new commit object in the repository, modifying the git history and refs. While git commits are technically recoverable via git reflog, they are persistent modifications to the repository state and cannot be undone without additional git operations.
From the tool's definition Tool description states 'Commit staged changes' and 'auto-stage tracked modified files', which are write operations that modify the git repository state irreversibly at the commit level, though the actual file content may be recoverable via git history.
Attacks that exploit this kind of access
Commit staged changes. Set all:true to auto-stage tracked modified files first. Auto-saves context entry. It is categorised as a Write tool in the Context MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Context MCP server in PolicyLayer and add a rule for git_commit: 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 Context. Nothing to install.
git_commit 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 git_commit 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 git_commit. 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.
git_commit is provided by the Context MCP server (vibhasdutta/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 →