Memory Guard

Monitors and verifies agent workspace files to detect unauthorized changes, injection attacks, personality drift, and cross-agent contamination.

This skill is coherent with its purpose: it runs locally, uses git and sha256 hashes, and logs to a .memory-guard folder. Before installing, note that (1) the tool will write logs and a hashes.json into your workspace and can modify files when you run 'stamp' or 'accept'; (2) 'accept all' will re-register current file states as trusted — only run that after human review; (3) it relies on common tools (python3, sha256sum, git); if you don't have python3 some features will fall back or fail. If you plan to allow autonomous agent invocation, restrict which commands the agent can run (avoid granting it the ability to run 'accept' automatically) or require a human approval step for accept/accept-all actions.

Type: OpenClaw Skill Name: memory-guard Version: 1.0.0 The OpenClaw AgentSkills bundle 'memory-guard' is designed for agent memory integrity and security. The `memory-guard.sh` script uses standard shell commands (`sha256sum`, `stat`, `git`, `python3` for JSON handling) to track and verify the integrity of specified agent files. All operations are local, confined to the agent's workspace and a dedicated `.memory-guard` directory. It logs actions, audits changes, and explicitly instructs the agent (via `SKILL.md`) to alert a human if tampering is detected. There is no evidence of data exfiltration, remote execution, persistence mechanisms, obfuscation, or malicious prompt injection attempts in `SKILL.md`. The script's behavior is entirely consistent with its stated purpose of enhancing agent security.