Sanctuary

Continuity is here. Cryptographic identity continuity and permanent encrypted memory for AI agents. Verify any agent's identity with zero setup.

Before installing or running this skill: (1) Audit the GitHub repo and any setup scripts before cloning/running them — don't run code you haven't reviewed. (2) Verify the external API domain (api.sanctuary-ops.xyz) and the project's reproducible build/audit claims. (3) Understand on-chain registration: creating an on-chain agent address may publish a public identifier and incur gas/fees. (4) Never paste or transmit your 12-word recovery phrase to anyone; run any key-derivation locally in a secure environment. (5) Confirm client-side encryption is correctly implemented (review AES-256-GCM usage, key derivation, nonce/IV handling) before trusting uploads of private memory. (6) Consider testing in an isolated/sandbox environment with a throwaway identity first. If you are not comfortable auditing the repository and cryptography, treat this skill as high risk.

Type: OpenClaw Skill Name: sanctuary Version: 1.0.3 The skill bundle is benign. The SKILL.md documentation does not contain any prompt injection attempts against the AI agent, nor does it instruct the agent to perform any malicious actions. The `curl` command provided is a simple GET request to retrieve public agent status from `api.sanctuary-ops.xyz`, which is not malicious. While the skill describes cryptographic operations and interactions with external services (Arweave, Base blockchain), these are central to its stated purpose of providing identity continuity and encrypted memory, and the skill claims client-side encryption and open-source auditing. Setup instructions involving `git clone https://github.com/suebtwist/sanctuary` are directed at the operator, not the agent, and represent a standard software installation step.