Reset HEAD/index/worktree by mode. Hard reset requires confirm=true.
AI agents call git_reset to permanently remove resources in Git — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Git reset, particularly hard reset, irreversibly discards work and cannot be undone without recovery tools. While soft/mixed resets are reversible, the tool's capability to perform hard reset (which destructively overwrites the working tree) and the explicit confirmation requirement place it in the Destructive category. The high severity reflects that misuse could cause permanent loss of uncommitted code changes.
From the tool's definition Tool performs 'reset' operation that can 'Reset HEAD/index/worktree by mode' including hard reset. Hard reset is irreversible and can permanently discard uncommitted changes.
Attacks that exploit this kind of access
Reset HEAD/index/worktree by mode. Hard reset requires confirm=true. It is categorised as a Destructive tool in the Git MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Git MCP server in PolicyLayer and add a rule for git_reset: 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. Nothing to install.
git_reset is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the git_reset 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_reset. 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_reset is provided by the Git MCP server (selfagency/git-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 →