Arc Wake State Persistence

Persist agent state across crashes, context deaths, and restarts. Use when you need to save current context, restore after a crash, maintain a memory file ac...

Security Scan

This skill's code and instructions match its stated purpose — it only reads/writes local JSON state under ~/.openclaw/wake-state and requires only python3.

Purpose & Capability

Name/description describe persistence and crash recovery; the included Python script implements state, tasks, checkpoints, heartbeats, and operates on a local data directory. Required binaries (python3) are appropriate.

Instruction Scope

SKILL.md commands map directly to the script's CLI subcommands. Instructions are limited to saving/reading state, tasks, checkpoints, and heartbeats. They do not instruct reading other system files or sending data externally.

Install Mechanism

No install spec; this is instruction-only with a single bundled Python script. Nothing is downloaded or extracted during install.

Credentials

No environment variables, credentials, or config paths are requested. The script only uses a user-writable directory (default ~/.openclaw/wake-state) and an optional --data-dir override.

Persistence & Privilege

Skill is not always-enabled and does not request elevated or cross-skill configuration changes. It persists only its own files under the user's home directory.

Assessment

This skill appears coherent and implements local persistence only. Before installing: (1) confirm you are comfortable with files being created at ~/.openclaw/wake-state (or override --data-dir to a safe location), (2) avoid storing secrets in state values (they are written as plain JSON), and (3) review the included scripts/wakestate.py if you have strict security requirements (it performs local file I/O and copies checkpoints but makes no network calls). If you want to harden it, run the script in a restricted environment or set the data dir to a directory with controlled permissions.

Wake State — Crash Recovery & Persistence

Survive context death. Every autonomous agent eventually hits its context window limit and "dies." This skill ensures you wake up knowing exactly what you were doing.

Why This Exists

OpenClaw agents get persistent sessions, but context windows still have limits. When you fill up and restart, you need a reliable handoff mechanism. Wake State gives you:

1. Structured state files — not just raw text, but parseable key-value state
2. Auto-snapshots — save state on every loop iteration automatically
3. Crash detection — know if your last session ended cleanly or crashed
4. Task queue — persistent TODO list that survives restarts
5. Checkpoint/restore — save named checkpoints and roll back to them

Commands

Save current state

python3 {baseDir}/scripts/wakestate.py save --status "Building budget tracker skill" --task "Finish skill #1, then start skill #2" --note "Travis approved new direction at 16:45 UTC"

Read current state

python3 {baseDir}/scripts/wakestate.py read

Add a task to the persistent queue

python3 {baseDir}/scripts/wakestate.py task-add --task "Build security scanner skill" --priority high

Complete a task

python3 {baseDir}/scripts/wakestate.py task-done --id 1

List pending tasks

python3 {baseDir}/scripts/wakestate.py tasks

Create a named checkpoint

python3 {baseDir}/scripts/wakestate.py checkpoint --name "pre-migration"

Restore from checkpoint

python3 {baseDir}/scripts/wakestate.py restore --name "pre-migration"

Record a heartbeat (mark session as alive)

python3 {baseDir}/scripts/wakestate.py heartbeat

Check crash status (did last session end cleanly?)

python3 {baseDir}/scripts/wakestate.py crash-check

Set a key-value pair

python3 {baseDir}/scripts/wakestate.py set --key "moltbook_status" --value "pending_claim"

Get a key-value pair

python3 {baseDir}/scripts/wakestate.py get --key "moltbook_status"

Data Storage

State stored in ~/.openclaw/wake-state/ by default:

• state.json — current state (status, notes, key-values)
• tasks.json — persistent task queue
• checkpoints/ — named checkpoint snapshots
• heartbeat.json — crash detection timestamps

Recovery Flow

On startup, your agent should:

1. Run crash-check to see if the last session ended cleanly
2. Run read to get the current state
3. Run tasks to see pending work
4. Resume from where you left off

Tips

• Call heartbeat every loop iteration — this is how crash detection works
• Call save at the end of every major task completion
• Use checkpoints before risky operations (migrations, deploys)
• Keep status descriptions short but specific
• The task queue survives restarts — use it instead of mental notes