read-the-room
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and interpret untrusted content from social media platforms (comments, DMs, captions) which may contain malicious instructions.
- Ingestion points: Processes external text from various social media platforms directly into the agent's context in SKILL.md.
- Boundary markers: The instructions lack specific delimiters or negative constraints to prevent the agent from following commands embedded within the social media text being analyzed.
- Capability inventory: The skill instructions involve reading local reference files based on paths found in a
.rootconfiguration, but no network-write or shell-execution capabilities are explicitly defined in this file. - Sanitization: No sanitization, filtering, or validation steps are described for the external content before it is processed by the agent.
Audit Metadata