Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
molt-life-kernel is a Node.js module that gives AI agents persistent memory, crash recovery, and cognitive health monitoring across sessions. It implements five operators: an append-only ledger, snapshot-based rehydration, human-in-the-loop witness gates, heartbeat vitality signals, and Shannon entropy coherence checks.
Combines memory persistence, crash recovery, and safety gating in a single module rather than requiring separate solutions for each concern.
When to use it
- Preserving agent context across session restarts
- Recovering agent state after an unexpected crash
- Requiring human approval before an agent deletes or modifies critical data
- Detecting when an agent has drifted from expected behavior over time
- Building an immutable audit trail of all agent actions for compliance
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: set up an agent that survives crashes and requires approval before deleting data
- 1Import MoltLifeKernel and instantiate with a witnessCallback pointing to a human approval function
- 2Call kernel.append() to record the incoming request to the append-only ledger
- 3Call kernel.enforceCoherence() to verify the agent's context has not drifted
- 4Call kernel.witness() with the proposed delete action and risk score before executing
- 5Call kernel.getSnapshot() to save a recovery capsule; if a crash occurs, call kernel.rehydrate() with the saved capsule and ledger
Agent executes the session with full audit trail, coherence validation, human-approved destructive actions, and a recoverable snapshot stored in the workspace