fan-operations
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The analyzed file consists entirely of markdown-based strategic guidance and does not contain any executable scripts or configuration files.
- [PROMPT_INJECTION]: The skill outlines procedures for interacting with untrusted external data from social media users, identifying an indirect prompt injection attack surface.
- Ingestion points: User comments and direct messages as referenced in 'SKILL.md'.
- Boundary markers: Not present; there are no specific instructions to separate or ignore instructions embedded in user content.
- Capability inventory: No technical capabilities, tool calls, or network operations are defined in this specific skill file.
- Sanitization: Not present; no validation or filtering logic is provided for the incoming user data.
Audit Metadata