fal-vision

The fal-vision skill is a catalogue entry that advertises image analysis via fal.ai vision models, including object segmentation, detection, OCR, description, and visual question answering. It exists within Open Design so an agent can discover the capability during planning through triggers such as "object detection", "ocr image", "visual qa", or "segment".

As shown in the source, this is a thin curated pointer rather than a complete implementation. The SKILL.md summarizes what the skill does and lists trigger phrases, but the actual assets, scripts, and references reside in the fal.ai community skills repository. To run the full workflow, the user inspects the upstream README for exact paths and installs the upstream bundle into their active agent's skills directory. The entry is marked image mode in the image-generation category.

The intended users are developers and analysts who need to extract information from images, whether that is detecting or segmenting objects, running OCR to pull text, describing image contents, or answering questions about an image. Because the entry is a discovery stub curated from the fal.ai community team, users should treat it as a signpost and consult the upstream repository for the specific vision endpoints, parameters, and usage details.