self-reflection
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill reads from 'self-review.md' to process past lessons. This creates an ingestion surface for indirect prompt injection if the workspace file is modified by untrusted processes. 1. Ingestion: self-review.md. 2. Boundary markers: None. 3. Capability: jq, date. 4. Sanitization: None.
- [Persistence Mechanisms] (SAFE): Utilizes the OpenClaw heartbeat configuration to enable periodic execution. This is a documented framework feature and necessary for the skill's primary purpose.
Audit Metadata