Workspace Cleanup Policy with Claude Code: MCP Tool Boundary

A production playbook for workspace cleanup policy in cross-industry operations using Claude Code: mcp tool boundary, run-scoped inputs, logs, typed results, and artifacts.

Audience: Platform teams managing ephemeral execution

The problem

Platform teams managing ephemeral execution need workspace cleanup policy to run repeatedly against temporary files, terminal states, storage rules, and retention policy. In cross-industry operations, the pain is not one good answer; it is repeatability, auditability, exception handling, and evidence that survives handoff.

Implementation path

Expose only the MCP tools needed for workspace cleanup policy, validate tool arguments, keep credentials in the owning service, and log each call outside the sandbox.

Tradeoffs and failure modes

Narrow tool boundaries reduce agent flexibility, but make the integration reviewable and supportable. For workspace cleanup policy, the practical test is whether a second run can be debugged, retried, and consumed by a product without reading the raw agent transcript.

Tool policy

tool: workspace-cleanup-policy_lookup
agent: Claude Code
input_scope: /skill/.argo/inputs
credential_owner: broker
log_arguments: true
network_policy: allowlisted

Run this on Argo