Explore

Find agent skills by outcome

105,054 skills indexed with the new KISS metadata standard.

Showing 24 of 105,054Categories: Operations & Workflow, Coding & Debugging, Data, General, Creative

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

I want people to say

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

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

Macro Sweep: Identify the scene type

Micro Sweep: Scan for textures

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

dominant_hex_estimates: [#RRGGBB

scene_description: A comprehensive

// REPEAT for EVERY single object

accent_colors: [Color name 1

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

Plain white background

- 7.0-8.9: Questioning

- 9.0-10.0: Distant

- 3.0-4.9: Playful

- 1.0-2.9: Vulnerable

- **Triggers**: Intellectual loneliness

Characters must look exactly like the original How to Train Your Dragon films

Hiccup and Toothless appear most frequently

Fully visible heads

redesign

Correct original outfits and dragon designs from the movies

- Triggers: Intellectual loneliness