Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Clean documentation: simplified SKILL.md, removed references to deleted scripts, focused on existing features
What This Skill Does
ContextKeeper tracks project state across AI agent sessions using manual checkpoints. It provides two shell scripts for creating checkpoints and viewing a status dashboard. No background processes or daemons are involved.
Avoids the security risks of checkpoint tools that use PID files, command substitution, or background watchers.
When to use it
- Saving work context before ending a coding session
- Resuming an AI agent task after a break
- Tracking progress on a multi-step refactor
- Logging milestones during a long debugging session
- Reviewing what was last worked on in a project
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: save my current progress on the auth fix
- 1Runs ckpt.sh with a descriptive message
- 2Script validates and escapes the input
- 3Checkpoint is recorded with the current project state
- 4User can later run dashboard.sh to see all checkpoints
Checkpoint 'Fixed auth issue' saved and visible in the project dashboard