Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Add 50-line/2KB size limit, completed task compression to one-line summaries, auto-prune >3 days
What This Skill Does
Maintains a live task state file at memory/tasks.md so an AI agent can resume context after session resets or conversation compaction. Tracks task status, background process details (session IDs, PIDs, server, command), and results. The agent updates this file automatically — it is not user-invoked.
Because state is written to a file before the agent reports back, context survives session resets that would otherwise wipe any in-memory task awareness.
When to use it
- Resuming a multi-step deployment after a session disconnect
- Tracking a background SSH process running on a remote server
- Recovering task context after conversation compaction
- Monitoring parallel background jobs without losing state
- Handing off in-progress work between agent sessions
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Deploy the app to the staging server
- 1Writes a new task entry with status 🔄 进行中 to memory/tasks.md
- 2Starts background SSH session and records session ID, PID, server name, and command
- 3Updates notes in the task entry after each deployment step
- 4Marks task ✅ 完成 and records result URL before reporting back to user
Deployment complete; task state persisted in memory/tasks.md with result link, resumable if session resets