A persistence layer for AI agents to maintain state, tasks, and checkpoints across crashes and context window resets.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install arc-wake-state
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install arc-wake-state using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Wake State?
Wake State is a specialized persistence layer designed to overcome the inherent limitations of LLM context windows and unexpected process terminations. By providing a structured mechanism to save and restore context, this skill ensures that autonomous agents can pick up exactly where they left off, effectively waking up with full awareness of their previous progress and pending objectives.
Integrating this skill is essential for building robust Openclaw Skills that require long-running autonomous workflows. It provides a reliable handoff mechanism that transforms ephemeral agent sessions into persistent, stateful operations that can survive context death and system restarts.
Wake State Use Cases
- Recovering progress after an agent hits a context window limit and requires a session restart.
- Maintaining a persistent TODO list that survives across multiple agent execution cycles.
- Implementing safety nets via named checkpoints before executing high-risk code migrations or deployments.
- Detecting system crashes using heartbeat monitoring to trigger automated recovery flows in Openclaw Skills.
How Wake State Works
- The agent records a heartbeat at every loop iteration to signal a healthy running state and facilitate crash detection.
- Progress and context are captured using structured state files and a persistent task queue stored on the local filesystem.
- Before performing risky operations, the agent creates a named checkpoint to allow for rapid rollback if an error occurs.
- Upon a restart or crash recovery, the agent executes a crash-check and reads the last saved state and task list.
- The agent resumes the workflow seamlessly based on the retrieved metadata and status reports.
Wake State Setup
To utilize Wake State within your environment, ensure you have Python 3 installed on a Linux or macOS system. The skill manages its own directory structure at ~/.openclaw/wake-state/ automatically.
Save your first state snapshot with:
python3 {baseDir}/scripts/wakestate.py save --status "Initial setup" --task "Bootstrap agent" --note "Starting persistence layer"
Wake State Data Schema & Taxonomy
The skill organizes its persistence data within the ~/.openclaw/wake-state/ directory using the following structure:
| File/Directory | Purpose |
|---|---|
state.json |
Stores current status, notes, and arbitrary key-value metadata. |
tasks.json |
A persistent queue for managing agent objectives and priorities. |
checkpoints/ |
Contains named JSON snapshots for versioned state restoration. |
heartbeat.json |
Stores timestamps used to determine if the previous session crashed. |
Wake State Advanced Features
- Automated crash detection that distinguishes between clean exits and process failures.
- Granular key-value storage for maintaining custom agent variables across restarts.
- Priority-based persistent task management for complex multi-step objectives.
- Named checkpointing and restoration to provide a robust undo mechanism for autonomous Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Trust Verifier
A security auditing tool that assesses the provenance, integrity, and publisher reputation of Openclaw Skills to ensure safe installation and execution.
Skill Scanner
A comprehensive security auditing tool designed to detect credential stealers, data exfiltration, and malicious code within Openclaw Skills.
Skill Health Monitor
A proactive monitoring utility designed to track performance drift, error rates, and resource usage for deployed AI skills.
Agent GitOps
Automated deployment, rollback, and version management for AI agent workflows and skills.
Warm Wake
A specialized grounding tool that restores an AI agent's sense of self and long-term memory before it resumes task execution.
Mission Claw Activity Logging
A command-line tool for logging AI agent activities, tracking token usage, and monitoring project progress through a local dashboard.
METADATA
Featured*
