
AI coding agent for terminal
Plandex is an open-source, terminal-based AI coding agent designed for planning and executing large coding tasks that span many steps and touch dozens of files. Built for real-world development workflows, Plandex combines multi-model AI with version control, sandbox review, and automated debugging to handle complex software engineering tasks.
Plandex handles up to 2M tokens of context directly and can index directories with 20M+ tokens using tree-sitter project maps across 30+ programming languages. A cumulative diff review sandbox keeps AI-generated changes separate from project files until approved, while command execution is controlled for easy rollback and debugging. The platform supports models from Anthropic, OpenAI, Google, and open-source providers, with context caching across all to reduce costs and latency. Full version control tracks every plan update with branching for exploring multiple paths.
Plandex is built for experienced developers working on large, complex codebases who need an AI assistant capable of handling multi-file, multi-step tasks. It particularly benefits developers who prefer terminal-based workflows and want fine-grained control over AI-generated changes through sandbox review and version control.
Install Plandex with the one-line, zero-dependency CLI installer from plandex.ai. You can self-host the server using Docker or use cloud hosting for convenience. Configure your preferred AI model providers (Anthropic, OpenAI, Google) and point Plandex at your project directory. Use natural language commands to describe tasks, review changes in the diff sandbox, and apply approved changes to your codebase.
Pricing & Accessibility: Plandex is completely free and open source. You bring your own API keys for the AI models you want to use (Anthropic, OpenAI, Google, etc.), so the only cost is your model API usage. Self-hosting with Docker is available for teams wanting full control, and cloud hosting is offered for convenience.
Why Consider Plandex: Plandex is the most capable open-source AI coding agent for large projects, offering unmatched context capacity (2M+ tokens), a unique diff review sandbox, multi-model support, and full version control for AI-generated changes in a terminal-based workflow.
Planning and executing large multi-file coding tasks, refactoring complex codebases, building features that span dozens of files, automated debugging of builds and tests, exploring multiple implementation approaches with branching
Free
Free tier: Unlimited (bring your own API keys)