onboard

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to collect API keys/service tokens from the user and "write the values into" a .env file (i.e., embed secrets into generated files), which requires the LLM to handle and output secret values verbatim despite a partial mitigation to avoid echoing them in full.

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text when the user provides a YouTube URL/“latest” and the workflow fetches that video’s transcript/content for processing (via yt-dlp and optionally Apify/youtube-transcript), which is not first-party authored.