Plan experimental auto-cleanup without terminating anything, even when auto-cleanup is disabled.
AI agents call auto_cleanup_dryrun to retrieve information from Clean Process Ended without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
A dry-run that plans cleanup without actually executing any termination is purely a read/inspection operation. No processes are killed, no data is modified, and the description explicitly guarantees no side effects even when auto-cleanup would normally be disabled.
From the tool's definition 'Plan experimental auto-cleanup without terminating anything' — explicitly states no termination occurs; it is a dry-run that only plans actions.
Attacks that exploit this kind of access
Plan experimental auto-cleanup without terminating anything, even when auto-cleanup is disabled. It is categorised as a Read tool in the Clean Process Ended MCP Server, which means it retrieves data without modifying state.
Register the Clean Process Ended MCP server in PolicyLayer and add a rule for auto_cleanup_dryrun: 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 Clean Process Ended. Nothing to install.
auto_cleanup_dryrun is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the auto_cleanup_dryrun 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 auto_cleanup_dryrun. 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.
auto_cleanup_dryrun is provided by the Clean Process Ended MCP server (marcelocaporale/clean-process-ended). 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 →