video-creation-suite

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

• Third-party content exposure detected (low risk: 0.10). The skill workflow involves extracting frames from a user-provided local video and analyzing them using an external Coze Bot API, which processes images rather than ingesting outsider-authored free text into a monitored channel.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill executes HTTP requests to download remote image assets and invokes external API endpoints during runtime without enforced provenance or signature checks, introducing an unverifiable runtime external dependency.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). Upon analyzing the provided skill documentation and scripts, a hardcoded API key was identified in scripts/coze_bot_client.py:309: api_key = "cztei_qHZQ0A5OSJjsmfZWmVb8bqu2BTbtB240YGbDYLhZpsIr8jER4aL4Aevyii8rnKfNs"

This string has high entropy, uses a realistic prefix ("cztei_"), and is directly supplied as a fallback default credential in the script rather than being a documentation placeholder (such as YOUR_API_KEY). Therefore, it represents an active/hardcoded secret leakage.