Remove Engineering Harness artifacts from the project. Deletes the .harness/ directory recursively. CHANGELOG.md and .github/* are intentionally preserved (external tool conventions). Set dry_run=true to preview the file list without deleting.
AI agents call harness_uninstall to permanently remove resources in Harness Engineering MCP — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
This tool performs irreversible deletion of the .harness/ directory and all its contents. While the dry_run flag allows preview without deletion, the primary function is destructive. The blast radius is high because removing project governance artifacts could break engineering workflows, CI/CD configurations, and policy enforcement.
From the tool's definition The tool description explicitly states it 'Deletes the .harness/ directory recursively' and 'Remove Engineering Harness artifacts from the project.' The recursive deletion of a directory is irreversible data destruction.
Attacks that exploit this kind of access
Remove Engineering Harness artifacts from the project. Deletes the .harness/ directory recursively. CHANGELOG.md and .github/* are intentionally preserved (external tool conventions). Set dry_run=true to preview the file list without deleting. It is categorised as a Destructive tool in the Harness Engineering MCP MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Harness Engineering MCP server in PolicyLayer and add a rule for harness_uninstall: 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 Harness Engineering MCP. Nothing to install.
harness_uninstall 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 harness_uninstall 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 harness_uninstall. 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.
harness_uninstall is provided by the Harness Engineering MCP server (movebrickschi/harness-engineering-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 →