bailian-multimodal-skills

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly shows and instructs placing an API key into command-line arguments and echoing it into a config file (e.g., --api-key and echo "sk-..."), which encourages embedding secret values verbatim in commands and outputs, creating exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). The skill accepts and fetches arbitrary external media URLs (e.g., --input-audio in run_asr, --input-images, --img-url, --reference-urls, --audio-url) and uses requests.get/_to_file_url to download or pass that untrusted public content into model APIs for transcription/generation, so third‑party content is ingested at runtime and can influence model behavior.