Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A file-based task management system where each task is a folder that moves through workflow stages: pending, in_progress, review, blocked, done, rejected, and implemented. Each stage has a defined set of documents and signal files that track status, assignment, and verification. Completion requires documented verification before a task can be marked done.
Using folders and signal files instead of a database or issue tracker keeps task state readable by any tool, agent, or human without needing a running service.
When to use it
- Creating a new feature task with objectives and acceptance criteria
- Moving a blocked task back to in_progress after a dependency resolves
- Submitting finished work to review with an implementation summary
- Recording why a task was rejected and what alternative was chosen
- Tracking which agent is assigned to a task and when it started
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Create a task for adding fuzzy search to the global search component
- 1Create folder docs/todo/pending/fuzzy_search/
- 2Write task.md with Summary, Objectives, and Acceptance Criteria sections
- 3Write context.md explaining current search implementation and why MiniSearch was chosen
- 4Create .priority file with content 'high'
- 5Create .version file with JSON schema version and created timestamp
Task folder at docs/todo/pending/fuzzy_search/ with task.md, context.md, .priority, and .version, ready to be picked up and moved to in_progress