local-agent-senses

Scheme0/local-agent-senses

★ 1 stars Python AI/LLM Updated today

Local multimodal vision & speech for text-only LLM agents (Codex, Claude Code, Cursor, Cline, Gemini CLI, and similar). OCR, image/video understanding, and audio transcription run entirely on your machine via Ollama + FunASR, with a CLI and MCP server.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "local-agent-senses": {
      "command": "uvx",
      "args": [
        "local-agent-senses"
      ]
    }
  }
}

Or install with pip: pip install local-agent-senses

README Excerpt

Local vision, video and speech tools for text-only LLM agents. The default backend is Ollama on your own machine; media is not uploaded unless you explicitly configure an OpenAI-compatible endpoint. 中文说明见 [README.zh-CN.md](README.zh-CN.md)。中文版本请见 [README.zh-CN.md](README.zh-CN.md)。 - Image understanding and verbatim OCR/transcription.

Tools (10)

VISION_DIRECT_URL_STREAMVISION_MAX_DOWNLOAD_MBVISION_MAX_DURATION_HVISION_MAX_IMAGE_MBVISION_MAX_PDF_PAGESVISION_MAX_STDIN_MBVISION_MCP_CACHEVISION_PDF_DPIVISION_SERVICE_EXECUTION_TIMEOUTVISION_SERVICE_QUEUE_TIMEOUT

local-agent-senses

Quick Install

README Excerpt

Tools (10)

Related AI/LLM Servers