gemini-cli
> Gemini Rust Suite 🦀: A powerful, modular Rust toolkit for interacting with Google Gemini. Features a feature-rich CLI, persistent semantic memory (LanceDB), and extensible tool integration via the Model Context Protocol (MCP).
claude mcp add --transport stdio frostdev-ops-gemini-cli gemini-cli-bin --command-mcp
How to use
Gemini Rust Suite exposes a CLI (gemini-cli) that includes built-in MCP servers for filesystem operations, command execution, and a memory store. These MCP servers let you extend Gemini’s capabilities by connecting external tools and services via the MCP protocol, enabling tool invocation from prompts and managing contextual memory. You can run the built-in MCP servers directly from the CLI using flags, or run the gemini-cli as a client to interact with an MCP host daemon (mcp-hostd) or an embedded MCP host. The suite also provides persistent memory integration and prompt enhancement to keep context across sessions. To use the MCP tools, start the Gemini CLI with the desired server flags or run the corresponding mcp servers through the mcp-hostd daemon, then issue tool calls from your prompts and Gemini will translate them into MCP tool executions.
How to install
Prerequisites:
- Rust toolchain (install from https://rustup.rs/)
- Basic Git and shell access
- Optional: Gemini API key for full Gemini API features
Manual installation steps:
-
Clone the repository (if not already): git clone https://github.com/your-username/gemini-rust-suite cd gemini-rust-suite
-
Build the release binaries: cargo build --release --bin gemini-cli-bin --bin mcp-hostd
-
Install binaries to a location in your PATH: cp target/release/gemini-cli-bin ~/.local/bin/ cp target/release/mcp-hostd ~/.local/bin/
-
Ensure ~/.local/bin is in your PATH. Then you can run the CLI and MCP host: gemini-cli-bin --filesystem-mcp mcp-hostd &
-
Optional: Use the provided install script from the project root to automate setup (if available): ./install.sh This will build, install binaries, and configure shell helpers.
Notes:
- The built-in MCP servers are invoked via CLI flags (e.g., --filesystem-mcp, --command-mcp, --memory-store-mcp). See repository docs for exact usage.
- After installation, reload your shell to enable the gemini wrapper function and, on Zsh, the mcpd helper.
Additional notes
Tips and known considerations:
- The MCP servers (filesystem, command, memory store) are built-in helpers. Use them to extend Gemini’s capabilities without external tooling.
- Ensure your GEMINI_API_KEY (if required) is configured in your environment to access Gemini APIs.
- If you plan to run mcp-hostd as a daemon, familiarize yourself with the mcpd controls (start/stop/status/logs).
- The CLI aims to provide session history across commands via a wrapper; ensure your shell config is reloaded after installation.
- If you encounter path issues, verify that the binaries (gemini-cli-bin and mcp-hostd) are executable and present in PATH.
Related MCP Servers
gemini-cli
An open-source AI agent that brings the power of Gemini directly into your terminal.
ai-trader
Backtrader-powered backtesting framework for algorithmic trading, featuring 20+ strategies, multi-market support, CLI tools, and an integrated MCP server for professional traders.
bridle
TUI / CLI config manager for agentic harnesses (Amp, Claude Code, Opencode, Goose, Copilot CLI, Crush, Droid)
sudocode
Lightweight agent orchestration dev tool that lives in your repo
mcp-discovery
A command-line tool written in Rust for discovering and documenting MCP Server capabilities.
architect
A powerful, self-extending MCP server for dynamic AI tool orchestration. Features sandboxed JS execution, capability-based security, automated rate limiting, marketplace integration, and a built-in monitoring dashboard. Built for the Model Context Protocol (MCP).