social
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a systematic framework for ingesting and analyzing untrusted data from external social media platforms, which could facilitate indirect prompt injection attacks.
- Ingestion points: The 'Reverse Engineering' framework in
references/reverse-engineering.mdinstructs the agent to collect and analyze 500-1000+ posts from external creators for analysis and pattern extraction. - Boundary markers: The instructions do not define clear boundary markers or provide warnings to the agent to disregard instructions that may be embedded within the scraped social media content.
- Capability inventory: The skill is primarily focused on content generation and strategy (writing posts, threads, and video scripts). No high-privilege capabilities such as file system modification, network exfiltration commands, or subprocess execution were identified in the provided files.
- Sanitization: There are no visible mechanisms for sanitizing or filtering instructions that might be present in the external content being processed during the 'Reverse Engineering' workflow.
Audit Metadata