blnk
@frgmt0
blnk について
概要
What is blnk?
blnk is a terminal-based chat application that enables seamless interaction with multiple AI providers, including OpenAI, Anthropic, and Google Gemini. It is designed for developers and terminal users who want a rich CLI experience with markdown formatting and MCP integration.
How to use blnk?
Install Python 3.13+, clone the repository, and run ./install.sh. Start blnk with blnk, then run /setup to configure API providers and keys. Use commands like /use, /switch-model, and /tools to interact with providers and MCP servers. Configuration lives in ~/.blnk/config/.
Key features of blnk
- Multi-provider support (OpenAI, Anthropic, Google Gemini)
- Rich CLI interface with markdown formatting
- Dynamic model switching and API selection
- MCP integration for advanced reasoning
- Streaming responses with progress indicators
- Command system with help documentation
- YAML configuration with automatic JSON conversion
- Secure API key storage in the user’s home directory
Use cases of blnk
- Chat with different AI models (e.g., Claude, GPT, Gemini) from one terminal
- Switch between providers or models mid-session to compare responses
- Configure MCP servers to give AI agents access to external tools and reasoning
- Use streaming commands to see responses in real time with progress feedback
FAQ from blnk
What AI providers does blnk support?
blnk supports OpenAI, Anthropic, and Google Gemini, with the ability to switch between them at runtime.
How do I install blnk?
Clone the repository, then run ./install.sh. Requires Python 3.13 or higher.
How do I configure an MCP server?
Create a JSON file in ~/.blnk/config/mcp/ (e.g., default.json) with the server command, arguments, and environment variables. Multiple servers are supported.
How are API keys stored?
API keys are stored securely in the user’s home directory as part of the blnk configuration.
Is blnk free and open source?
Yes, blnk is licensed under the MIT license and is free to use.
「その他」の他のクライアント
mcp_ha_xiaozhi
shawn996小智官方服务器(虾哥)对接home assistant的mcp server

Prove AI
preuve.aiStartup research engine. Scores ideas with live market data. Only 17.5% pass.
Llama MCP Streamlit
nikunj2003AI assistant built with Streamlit, NVIDIA NIM (LLaMa 3.3:70B) / Ollama, and Model Control Protocol (MCP).
Dify Connect MCP
difybaseAnirudh's Chat
kamathA simple NextJS MCP client with sensible keybindings
コメント