clean
Runs dotnet clean to remove build outputs and returns structured results.
This record as markdown: /tools/io-github-dave-london-docker/clean.md
What clean does on Docker
AI agents call clean to permanently remove resources in Docker, typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Why clean is rated Critical
While the scope is limited to build artifacts rather than source code or data, the operation is irreversible and cannot be undone. An AI agent misusing this could cause loss of build state, requiring expensive recompilation. This exceeds Write (reversible modification) and qualifies as Destructive because deletion cannot be undone.
From the tool's definition Tool name 'clean' and description explicitly states it 'remove[s] build outputs' — a destructive operation that irreversibly deletes compiled binaries, intermediate artifacts, and cache files.
Attacks that exploit this kind of access
The rule that runs clean safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Docker, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For clean, this is the rule to start with:
clean is removed from the agent's tool list entirely, so the agent never calls it. The rest of the server keeps working.
The button opens the PolicyLayer dashboard: create your workspace, connect Docker, apply this rule, and every clean call is checked against it from then on.
Questions about clean
Runs dotnet clean to remove build outputs and returns structured results. It is categorised as a Destructive tool in the Docker MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Docker MCP server in PolicyLayer and add a rule for clean: 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 Docker. Nothing to install.
clean 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 clean 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 clean. 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.
clean is provided by the Docker MCP server (@paretools/docker). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Docker, and thousands of servers like it.
Across the catalogue