Resets repository to a specific commit or state
AI agents call git-reset to permanently remove resources in GitHub MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
git-reset is destructive because it can irreversibly discard or overwrite uncommitted changes and rewrite commit history depending on the mode used. While not as severe as force-pushing to shared branches, it represents data loss that cannot be automatically recovered. An AI agent misusing this without user consent could lose recent work.
From the tool's definition Tool name is 'git-reset' and description states it 'Resets repository to a specific commit or state'. The git reset command can discard uncommitted changes and move HEAD to prior commits, effectively undoing work.
Attacks that exploit this kind of access
Resets repository to a specific commit or state. It is categorised as a Destructive tool in the GitHub MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the GitHub MCP Server 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 GitHub MCP Server. 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 GitHub MCP Server MCP server (jungchihoon/github-mcp-server). 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 →