Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
A protocol for consent-governed cryptographic memory management in relational systems. Combines sharding, differential privacy, and multi-party quorum to protect stored context under adversarial pressure. Participants retain sovereign erasure rights at all times.
Most memory architectures depend on operator trustworthiness for confidentiality; this protocol is designed to remain safe even when operator access becomes adversarial.
When to use it
- Protecting sensitive relational data from jurisdiction seizure or operator capture
- Triggering automatic dormancy when harm spikes or coercion attempts are detected
- Running governance conformance tests before promoting a memory system to production
- Composting stored memory permanently at participant request
- Distributing memory shards across jurisdictions to prevent single-point concentration
Example Workflow
Here's how your AI assistant might use this skill in practice.
User requests erasure of their stored relational memory
- 1Validate the erasure request as sovereign and irreversible, requiring no quorum consensus
- 2Trigger key rotation and shard invalidation across all jurisdictions
- 3Propagate revocation network-wide within the 120-second SLA
- 4Generate cryptographically attestable invalidation receipts per hop
- 5Log the erasure event with a ZK proof in the audit trail, without identity linkage
All shards composted, revocation confirmed with cryptographic receipts, audit trace preserved without identity-linked content