c-level-skills
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill architecture described in the index facilitates the ingestion of external data, which creates a potential surface for indirect prompt injection.
- Ingestion points: The
cs-onboardskill (founder interview data) and theboard-meetingskill (independent contributions from multiple agents). - Boundary markers: Absent; the index does not specify the use of delimiters or 'ignore' instructions for embedded content.
- Capability inventory: The skill bundle maintains write access to sensitive local configuration paths including
~/.claude/company-context.mdand~/.claude/decisions/for persistence. - Sanitization: Absent; the documentation does not describe any validation or escaping of external inputs before they are processed by the advisors.
- [NO_CODE]: This file functions exclusively as a 'Bundle Index' and router. It does not contain any executable code, scripts, or tool definitions itself, although it references 37 external stdlib-only Python tools and 33 sibling skills.
Audit Metadata