cs-onboard

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The required workflow for /cs:setup and /cs:update collects the founder’s free-form interview answers and writes them into ~/.claude/company-context.md, which is then used as LLM context—this is outsider-authored text (the founder/interviewee) entering the agent’s prompt context.