Explore

Find agent skills by outcome

115,137 skills indexed with the new KISS metadata standard.

Showing 24 of 115,137Categories: Research & Learning, Creative, General, Data, Coding & Debugging

Vision-to-json

This is a request for a System Instruction (or "Meta-Prompt") that you can use to configure a Gemini Gem. This prompt is designed to force the model into a hyper-analytical mode where it prioritizes c...

Find agent skills by outcome

Vision-to-json

scene_description: A comprehensive

// REPEAT for EVERY single object

Micro Sweep: Scan for textures

CORE DIRECTIVEDo not summarize. Do not offer high-level overviews unless nested within the global context. You must capture 100% of the visual data available in the image. If a detail exists in pixels

accent_colors: [Color name 1

Don’t give me boring

CORE DIRECTIVEDo not summarize. Do not offer high-level overviews unless nested within the global context. You must capture 100% of the visual data available in the image. If a detail exists in pixels

Before generating the final JSON

Macro Sweep: Identify the scene type

Relationship Sweep: Map the spatial and semantic connections between objects (e.g.

Use my voice: smart

Micro Sweep: Scan for textures

Macro Sweep: Identify the scene type

Before generating the final JSON

You must return ONLY a single valid JSON object. Do not include markdown fencing (like ```json) or conversational filler before/after. Use the following schema structure

. I want to promote the product in a way that feels authentic

I’m good at showing BTS

● I post on Instagram (Reels + carousels + Single image)

Sticker

● CTA (if any): [Subtle or direct

- 9.0-10.0: Distant

- 7.0-8.9: Questioning

- 3.0-4.9: Playful