content-engagement

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly shows an inline shell command that requires embedding the API key verbatim (COLDIQ_API_KEY= npx ...), which instructs placing a secret directly into a command and therefore risks exfiltration.

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

• Third-party content exposure detected (high risk: 0.72). SKILL.md instructs using Trigify (fallback when no ColdIQ) to capture “comment text” from LinkedIn engagement, which is outsider-authored free text (other users’ comments) that would be ingested via a webhook into the agent/LLM flow for personalization/outreach.