code-assistant
by stippi·★ 163·Score 48
Rust-based AI coding assistant with GUI, CLI, MCP and ACP modes offering autonomous code analysis and modification.
Overview
Code Assistant is a sophisticated AI-powered coding tool built in Rust that provides multiple interfaces for autonomous code analysis and modification. It features adaptive tool execution that supports different LLM providers and their unique function calling capabilities. The tool offers real-time streaming with smart filtering to prevent unsafe tool combinations and session-based project management with persistent state. It supports both Model Context Protocol (MCP) for integration with Claude Desktop and Agent Client Protocol (ACP) for editor integration like Zed. The application includes features like session compension when approaching context limits and automatic formatting of modified files.
Try asking AI
After installing, here are 5 things you can ask your AI assistant:
When to choose this
Choose this when you need a sophisticated AI coding assistant with multiple LLM provider support, especially if you're working on Rust projects or need format-on-save functionality.
When NOT to choose this
Avoid this if you need simple code completion without session state or prefer a more lightweight solution without GUI dependencies.
Comparable tools
Installation
Installation
# Install Rust tool chain via rustup
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
# Clone and build
git clone https://github.com/stippi/code-assistant
cd code-assistant
cargo build --releaseClaude Desktop Integration
Configure in Claude Desktop settings (**Developer** tab → **Edit Config**):
{
"mcpServers": {
"code-assistant": {
"command": "/path/to/code-assistant/target/release/code-assistant",
"args": ["server"],
"env": {
"SHELL": "/bin/zsh"
}
}
}
}FAQ
- Which models does Code Assistant support?
- Code Assistant supports multiple LLM providers including Anthropic, OpenAI, Ollama, SAP AI Core, Vertex AI, Groq, Cerebras, MistralAI, and OpenRouter. Models are configured in a JSON file where you can specify provider, model ID, and various parameters.
- How does it handle different tool syntax formats?
- The tool offers three syntax modes: native (provider's built-in tool calling), XML-style tags for streaming parameters, and triple-caret blocks for token efficiency. It automatically adapts based on the target LLM's capabilities.
Compare code-assistant with
Last updated · Auto-generated from public README + GitHub signals.