skill-14-social-reputation
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from external social protocols (Lens API, Farcaster Hub), which creates a surface for Indirect Prompt Injection. Malicious social posts or profile metadata could contain instructions intended to influence the agent's behavior.
- Ingestion points: Lens Protocol publications (posts, comments) and Farcaster casts.
- Boundary markers: None specified in the instructions for isolating external content.
- Capability inventory: The skill allows for posting on behalf of users, following accounts, and collecting NFTs, which could be abused if the agent follows malicious instructions in processed data.
- Sanitization: The skill recommends filtering spam and low-reputation accounts, which provides a partial defense-in-depth against automated malicious content.
Audit Metadata