working-with-sergey
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and review untrusted user-provided content, such as draft DMs, invites, and proposals, to ensure they adhere to specific communication norms.
- Ingestion points: User-provided drafts processed during the "Review" mode as described in
SKILL.md. - Boundary markers: The skill instructions do not specify explicit delimiters or "ignore instructions" wrappers for the ingested content.
- Capability inventory: The skill consists entirely of Markdown instructions and lacks any executable capabilities, tools, or network access.
- Sanitization: There is no evidence of sanitization or filtering for the input text, though the lack of capabilities prevents this from being a functional vulnerability.
- [NO_CODE]: The skill consists exclusively of Markdown documentation and instructional tables. No scripts (Python, JavaScript, shell) or configuration files that trigger code execution were found in the provided files.
Audit Metadata