mpi-cleanup

Identify active, orphaned, and archived MPI Kanban artifacts for cleanup approval.

1|Updated May 13, 2026
One-click install
npx skills add https://github.com/MadPonyInteractive/mpi-kanban --skill mpi-cleanup
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mpi-cleanup
Source: https://github.com/MadPonyInteractive/mpi-kanban/tree/main/skills/mpi-cleanup
Command: npx skills add https://github.com/MadPonyInteractive/mpi-kanban --skill mpi-cleanup

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Conservatively cleans up MPI workflow artifacts and stale coordination data to prevent clutter in task workspaces, archived plans, handoffs, and state folders, while preserving active work and knowledge.

Core Features & Use Cases

  • Proposes cleanup actions by scanning active boards, tasks, handoffs, and state, then surfaces safe options for archiving or deletion with explicit user approval.
  • Maintains active project knowledge (project-profile.md and project-knowledge-index.md) and never deletes essential knowledge without confirmation.
  • Suitable for end-of-session wrap-ups, ongoing maintenance, and drift cleanup in multi-agent MPI sessions.

Quick Start

Ask your agent to run MPI cleanup and approve the proposed actions when prompted.

Frequently Asked Questions about mpi-cleanup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I clean up stale Kanban tasks and handoffs without losing active project state?

Kanban cleanup identifies active, orphaned, and archived artifacts across boards, tasks, and handoffs, then proposes safe archiving or deletion actions for explicit user approval before any files are mutated.

How does conservative workspace cleanup handle essential knowledge files like project-profile.md?

Conservative workspace cleanup maintains active project knowledge by preserving project-profile.md and project-knowledge-index.md, ensuring essential knowledge is never deleted without explicit confirmation during the final approval prompt.

What is the best way to archive orphaned plans and coordination data in a multi-agent session?

Archiving orphaned plans and coordination data is done by scanning .agents/mpi-kanban/ and docs/plans/ to surface stale artifacts, presenting safe archive or deletion options that await your explicit confirmation before execution.

Can I use this workspace cleanup for end-of-session wrap-ups and ongoing drift maintenance?

Yes, workspace cleanup is suitable for end-of-session wrap-ups, ongoing maintenance, and drift cleanup in multi-agent sessions, scanning state and handoff files to propose conservative archiving actions with user-approved execution.

Does the cleanup process automatically delete files from the .agents/mpi-kanban/ directory?

No, the cleanup process does not automatically delete files; it proposes actions by analyzing the .agents/mpi-kanban/ directory and related knowledge indexes, waiting for explicit user approval before performing any mutation or deletion.

When should I avoid running a conservative state-management cleanup on my workspace?

You should avoid running state-management cleanup when you have unsaved or uncommitted active work, as the process targets drift and stale artifacts but requires explicit approval to prevent accidental loss of essential coordination data.