meeting-prep

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes an explicit curl example that requires embedding a Slack bot token in an Authorization header (and lists a Slack bot token as a prerequisite), which instructs the agent to place a secret verbatim into a command/request.

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

• Third-party content exposure detected (high risk: 0.90). SKILL.md explicitly instructs the agent to perform web searches and extract LinkedIn profiles, company news, and Crunchbase data (e.g., "Search: '[attendee name] [company] LinkedIn'" and "Search: '[company] recent news'"), so the agent will fetch and interpret untrusted public third‑party content that can materially influence briefs and follow-up actions.