Terminai (Termin.AI) is an open-source terminal process manager and AI agent harness developed by emosenkis. Built on top of the mature mprocs process runner codebase, Terminai combines multi-process terminal orchestration with native LLM assistant capabilities directly inside the terminal user interface.
mprocs-Based Process Management — Runs and coordinates multiple simultaneous background and foreground terminal processes using portable-pty. - Ratatui Terminal UI — Delivers a responsive, split-pane TUI for monitoring process outputs, input handling, and interacting with AI side-by-side. - Integrated AI Process Panel — Hosts dedicated conversational LLM sessions within the terminal workspace to assist with running commands. - Terminal I/O Capture — Observes terminal output and process state to provide context-aware LLM command suggestions and diagnostics. - Built-in Privacy Filtering — Includes dedicated privacy filtering modules (src/privacy/) to sanitize terminal data prior to sending it to model APIs. - Remote Control & Copy Mode — Supports native remote control APIs, keyboard navigation, and scrollback copy buffers inherited from mprocs.
Use cases
Use Case: Debugging Multi-Service Development Stacks Developers running frontend dev servers, backend APIs, and database migrations simultaneously in a single terminal can use the AI assistant pane to diagnose crash logs or failed builds across running processes without context-switching to a browser.
Use Case: Interactive Command Generation Engineers working with unfamiliar CLI tooling or deployment scripts can prompt the integrated AI chat process to draft, refine, and parse CLI commands directly within their active terminal session.
Use Case: Context-Aware Log Analysis Terminal users can capture active process I/O and pass filtered logs through Terminai's privacy layer to analyze runtime exceptions, network timeouts, or configuration errors.
Who it is for
Terminai is intended for developers, DevOps engineers, and system administrators who already use terminal multiplexers or process runners (like tmux or mprocs) and want integrated AI assistance alongside running processes. It is less suitable for users looking for automated headless background agents or IDE-native GUI plugins.
Install guide
Installation Terminai is written in Rust and can be built from source using Cargo: bash git clone https://github.com/emosenkis/terminai.git cd terminai cargo build --release Move the compiled binary to your system $PATH: bash sudo cp target/release/terminai /usr/local/bin/
First Run Configure your model provider API key in your environment and set up any optional process configurations: bash export OPENAI_API_KEY="your-api-key-here" # or set ANTHROPIC_API_KEY depending on your LLM backend configuration Launch Terminai directly or pass a configuration file defining your processes: bash terminai
Example Session Navigate between your active process tabs and the AI assistant pane using the configured TUI hotkeys (default standard mprocs navigation). Focus the AI input box to ask questions regarding active terminal output: text AI> Analyze the error log from the backend process on port 8080 and suggest a fix.
GitHub Copilot CLI — GitHub Copilot CLI is a terminal-based AI coding assistant developed by GitHub. It integrates the Copilot coding agent runtime directly…
Aider — Aider is an open-source terminal-based AI pair programming tool created by Paul Gauthier and the Aider-AI community. It operates directly…
Agentty — Agentty is an open-source, high-performance native terminal-based AI coding agent written in C++ by 1ay1. Engineered specifically for speed and…
Amp — Amp (developed by AmpCode) is an autonomous coding agent platform and remote development harness designed to execute software engineering tasks…
Cline — Cline is an open-source autonomous coding agent runtime maintained by Cline Bot Inc. and an active open-source community. Available as…
Continue — Continue is an open-source AI code assistant and agent framework originally developed by the Continue team (continuedev) and recently acquired…