fal-upscale

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

• Third-party content exposure detected (high risk: 0.70). The skill is sourced from the public fal-ai-community/skills GitHub repository, and at runtime it would ingest the upstream workflow’s README/assets/scripts and any referenced community-provided text content (outsider-authored free text) into the agent’s LLM context during skill execution/planning.