fal-api

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

• Third-party content exposure detected (high risk: 0.70). The skill accepts arbitrary external URLs (e.g., the audio_url parameter in FalAPI.transcribe and the image_url parameter in FalAPI.generate_video) which are sent to the fal.ai queue endpoints (https://queue.fal.run/...) for processing and return transcripts/results that the agent will read, meaning untrusted, user-provided third-party content can be ingested and interpreted as part of the workflow.