tarot-reading
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill relies exclusively on local markdown files for its knowledge base and interpretation logic, ensuring no external data is fetched and no network requests are made during the reading process.
- [SAFE]: Persistence is limited to a local
MEMORY.mdfile used for storing user-specified preferences like favorite spreads. This is a standard state-management pattern and does not involve accessing sensitive system credentials or unauthorized file paths. - [SAFE]: The skill instructions contain robust safety constraints and prohibitions against making medical, legal, or fatalistic predictions, preventing the agent from providing inappropriate advice.
- [SAFE]: No obfuscated content, remote code execution patterns, or suspicious administrative commands were found in the provided instruction files or metadata.
Audit Metadata