x-repurpose
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content such as transcripts, articles, and call summaries, which inherently introduces a surface for indirect prompt injection. \n
- Ingestion points: User-provided text or transcript tool output (SKILL.md). \n
- Boundary markers: The instructions lack specific delimiters or negative constraints to prevent the agent from following instructions embedded within the source material. \n
- Capability inventory: The skill reads a local hooks file ('../x-post/hooks.json') and coordinates with other skills ('/x-post', '/x-thread', '/x-human'). \n
- Sanitization: No explicit validation or filtering of the source content is mentioned in the instructions. \n- [SAFE]: The skill's access to external resources is limited to reading a template file from a sibling directory, which is standard for inter-skill resource sharing in agent platforms and does not involve sensitive data exposure.
Audit Metadata