An Agentic, Multi-Model CLI Coding Assistant
npm install @indiccoder/mentis-cliEsc to instantly stop generation.
/resume) and checkpoints (/checkpoint).
grep tool.
bash
npm install -g @indiccoder/mentis-cli
`
$3
`bash
git clone https://github.com/CoderVLSI/Mentis-CLI.git
cd Mentis-CLI
npm install
npm run build
npm link
`
βοΈ Configuration
Start Mentis by typing:
`bash
mentis
`
$3
Type /model inside the CLI to launch the interactive configuration wizard:
1. Select Provider: Gemini, Ollama, or OpenAI.
2. Select Model: e.g., gemini-2.5-flash, llama3.
3. Enter Credentials: API Key (for cloud) or Base URL (for local).
Credentials are stored securely in ~/.mentisrc.
$3
Mentis includes web search capabilities for documentation and error solving. To enable it:
1. Get a free Tavily API key at https://tavily.com
2. Add to your .env file in your project directory:
`
TAVILY_API_KEY=your_key_here
`
Alternatively, you can use Exa via MCP:
`bash
export EXA_API_KEY=your_key_here
mentis
/mcp connect "Exa Search"
`
π Usage
$3
* /plan: Switch to high-level planning mode (Architecture, requirements).
* /build: Switch to code generation mode (Implementation).
$3
| Command | Description |
| :--- | :--- |
| /help | Show available commands |
| /model | Configure AI provider and model (Interactive) |
| /resume | Resume the last session |
| /checkpoint | Manage saved sessions (save, load, list) |
| /search | Search for code in the current directory |
| /mcp connect | Connect an MCP Server (e.g., npx -y @modelcontextprotocol/server-time) |
| /run | Execute a shell command |
| /commit [msg] | Stage and commit changes to Git |
| /add | Add a specific file to context |
| /drop | Remove a file from context |
| /clear | Clear chat history |
| /exit | Save and exit |
$3
`text
Mentis > /plan
[PLAN] > Analyze this project structure and suggest improvements.
Mentis > /build
[BUILD] > Implement the FolderManager class in src/utils.ts.
``