Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A file-based methodology for tracking structured work using plans stored in a folder hierarchy. Plans move through defined lifecycle states (draft, approved, in_progress, completed, archived) and contain proposal, design, and task documents. Supports both solo agents and multi-agent teams through parent-child plan delegation.
Keeping plans as versioned files in a workspace means any agent or human can inspect, resume, or delegate work without relying on shared databases or external project management tools.
When to use it
- Tracking a multi-phase feature build across agents
- Delegating a sub-task to another agent with linked plans
- Getting sign-off on a proposal before starting execution
- Reviewing what work is currently in progress at session start
- Organizing a project with separate specs, design decisions, and task checklists
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Create a plan to migrate our database schema
- 1Create plans/draft/2026-03-13-db-schema-migration/ folder
- 2Write proposal.md with problem statement, approach, and risks
- 3Write design.md with migration steps and dependencies
- 4Write tasks.md with phased checkboxes and owners
- 5Move folder to plans/approved/ after user approves the proposal
A structured plan folder ready for execution, with proposal, design, and task documents linked to the approval decision