Enterprise DNAEnterprise DNA
MMCP ServersDeveloperlow

metabolism-tools/workspace-metabolism

by Various

Govern what Claude Code, Codex, Aider and OpenClaw leave in your workspace: one JSON policy file, audit, recyclable clean, rollback, hash-chained audit trail.

MCP

metabolism-tools/workspace-metabolism

Added 21 Sept 2026

#ai-agents#ai-governance#audit#claude-code#cleanup#cli#developer-tools#dsh-plugin
Catalogue record, not a reviewed recommendation. This record can include automatically collected descriptions and setup details. Confirm important claims with the original source. Explore source-reviewed tools.

Overview

Workspace governance tool for AI coding agents. Uses a single JSON policy file to control, audit, and clean files left by Claude Code, Codex, Aider, and OpenClaw. Provides rollback and a hash-chained audit trail.

Best for

Best for
Developers using multiple AI coding agents who want centralized workspace cleanup and audit.

Use cases

  • Enforce cleanup policies for agent-generated files
  • Audit workspace changes with a hash-chained trail
  • Roll back unwanted files left by coding agents

How to use

Install

pip install workspace-metabolism

Tools exposed

  • retention_days
  • remote_authoritative
  • review_after

Clients listed in this record (not verified here)

Claude Code, Cursor

Notes

2 stars on GitHub. Last updated 2026-09-21. Licensed MIT.

Pros

  • Single JSON policy file covers multiple agents
  • Includes audit and rollback capabilities
  • Simple Python-based tooling

Cons

  • Very low GitHub star count (2) suggests limited community traction
  • Only supports the four listed agents
  • Requires Python setup and manual policy configuration
Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.