social-content
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill provides a framework for analyzing external social media posts, which constitutes a data ingestion surface for the agent.
- Ingestion points: The 'SCRAPE' step in 'references/reverse-engineering.md' directs the agent to collect and analyze 500+ external posts.
- Boundary markers: No instructions for delimiters or ignore-instructions are provided to isolate external content from system prompts.
- Capability inventory: The skill focus is limited to reasoning and text generation; no tool execution or network access is included in the skill files.
- Sanitization: There are no instructions for sanitizing or filtering malicious prompts that might be embedded in the ingested posts.
Audit Metadata