Back to directory
Frontendwuxiangru915/pi-vision-bridge
@wuxiangru/pi-vision-bridge
Give text-only models (DeepSeek, Llama, etc.) vision — describe_image tool + automatic image-to-text fallback, using any vision model configured in pi's model registry.
Suggested install command
npx skills add wuxiangru915/pi-vision-bridge/@wuxiangru/pi-vision-bridgeAlways 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 |