comment-funnel

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly tells the agent to ask for the user's API key and the documentation shows curl examples embedding "Authorization: Apikey YOUR_KEY" headers, which requires the LLM to accept and emit API keys verbatim in requests/commands.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md describes runtime calls to Upload-Post endpoints that return Instagram comment text (outsider-authored user-generated content) which the agent then “analyze[s] each comment for intent” and uses for DM personalization, so the LLM context can include that free text via the comment-reading/analyzing path.