n8n:loom-transcript

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

• Potentially malicious external URL detected (high risk: 1.00). The skill performs runtime fetches from https://www.loom.com/graphql and then downloads the returned source_url / captions_source_url transcripts which are injected into the agent's output/prompt, so external content directly controls the agent's produced instructions/output.