Agnes understand image

Skill oh-summy/skills/skills/agnes-understand-image

Cross-agent AI skills for Claude, Codex, OpenCode, Kimi Code, and more. Install all at once or pick only what you need.

Install

npx -y skills add oh-summy/skills --skill agnes-understand-image

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Analyze an image with Agnes AI and return structured information. Supports public image URLs, local image files, and JSON output.

SKILL.md

2.1 KB, 476 tokens by cl100k_base, as published. Nobody here has run it

Agnes Understand Image

Analyze an image with agnes-2.0-flash and return a structured understanding.

When to use

The user asks "what is in this image", "describe this screenshot", or "extract text/structure from this image".
The user provides an image file path or URL and wants a concise or JSON-formatted analysis.

Inputs

--image: Path to a local image file or a publicly accessible image URL.
--prompt: The analysis instruction, e.g. "Describe this image in one sentence" or "Extract all text as JSON".
--format: text, structured (default), or json.
--json: Alias for --format json.
--model: Optional model override. Default is agnes-2.0-flash.

Steps

Confirm the image source with the user if it is not already provided.

Run the helper script in this skill directory:

bash scripts/understand_image.sh --image <path-or-url> --prompt "<instruction>"

Return the model output to the user. If --json was used, validate that the output is valid JSON before presenting it.

Example

bash scripts/understand_image.sh \
  --image ./screenshot.png \
  --prompt "List the UI elements and their likely purpose as JSON" \
  --json

Output format

By default the model returns a structured analysis including:

Summary
Description
Subjects
Style
Color palette
Composition
Mood
Text in image
Metadata (dimensions, aspect ratio, format)

Use --format json to get machine-readable JSON. See references/output-schema.md for the full schema.

Templates / references

references/output-schema.md — structured output schema

API reference

Model: agnes-2.0-flash
Docs: https://agnes-ai.com/zh-Hans/docs/agnes-20-flash

Setup

Set your Agnes API key:

export AGNES_API_KEY=your_key_here

Or create a .env file from the repository root .env.example.

What ships with it: 3 files

6.5 KB alongside SKILL.md, 1 of them executable

references/

output-schema.md937 B

scripts/

understand_image.shruns5.5 KB

.env.example151 B