Back to directory
Docs & WritingKoreyer/easy-vision
easy-vision
A DeepSeek Harness tool plugin that lets a text-only agent 'see' local images: it reads an image file, auto-detects the real format, and returns (or writes to a Markdown file) a detailed text description produced by a separate OpenAI-compatible vision mod
Suggested install command
npx skills add Koreyer/easy-vision/easy-visionAlways inspect the linked repository and skill instructions before running commands. Skills are instructions; permissions and execution still matter.
Compatibility
Agent support matrix
3 supported
| Agent | Status |
|---|---|
| Claude Code | Supported |
| OpenCode | Not listed |
| Cursor | Supported |
| MCP | Not listed |
| GitHub Copilot | Not listed |
| Windsurf |