social-signal-bot
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill is entirely descriptive and does not contain any executable scripts, shell commands, or configuration files that could be run by an agent.
- [Data Exposure & Exfiltration] (SAFE): The instructions explicitly advise users to store secrets safely and mask tokens in logs, demonstrating a security-first approach to credential management.
- [Indirect Prompt Injection] (LOW): The 'Monitoring' section describes listening to mentions and replies from social media platforms. This is a potential surface for indirect prompt injection if an implementation were provided, as it involves processing untrusted external data.
- Ingestion points: Mentions and replies from X (Twitter) and Telegram.
- Boundary markers: Not applicable (documentation only).
- Capability inventory: None (no scripts provided).
- Sanitization: Mentions avoiding auto-engagement with scams, which is a form of manual/logic-based sanitization.
Audit Metadata