Logo
Deploy Now

SaaS Alternative

Devin Replit Agent Cursor

Stars

15,591

Forks

1,172

Watchers

106

Developer links

Plandex

With 15,500 GitHub stars and over 1,100 forks, Plandex delivers a terminal-based AI coding agent purpose-built for the complex, multi-file tasks that overwhelm single-file AI assistants. The Go-powered server maintains a cumulative diff review sandbox that quarantines all AI-generated changes from your project files until you explicitly approve them — enabling 20-file refactors where you cherry-pick good changes and reject bad ones without touching git. A 2M token effective context window loads only what each step requires, while tree-sitter project maps index repositories exceeding 20M tokens across 30+ programming languages, providing structural awareness of class hierarchies, function signatures, and import graphs without burning tokens on full file content. The configurable model pack system assigns different models to different roles — Claude for planning, GPT for coding, Gemini for summarization — supporting Anthropic, OpenAI, Google, OpenRouter, Azure OpenAI, AWS Bedrock, DeepSeek, Perplexity, and Ollama for local models. Full auto mode handles end-to-end autonomous workflows including high-level planning, context loading, implementation, terminal command execution, and automated debugging of both terminal and browser applications. The interactive REPL provides fuzzy auto-complete, version-controlled sandbox branching, rewind to any previous point, and Git integration for commit message generation. The Plandex Server exposes 60+ REST API endpoints for programmatic orchestration across organizations, projects, plans, and branches. Deploy via Docker Compose for self-hosted operation with your own API keys. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Plandex
Plandex
Plandex
Plandex

Benefits

  • 2M Token Context Window
  • Handles projects exceeding 20 million tokens using tree-sitter structural maps across 30+ languages, loading only relevant context per step to maximize model accuracy and minimize API costs.
  • Cumulative Diff Review Sandbox
  • Quarantines all AI-generated code changes in a version-controlled sandbox with branching, rewind, and selective application so you never commit untested modifications directly to your codebase.
  • Multi-Provider Model Flexibility
  • Assigns different AI models to planning, coding, and debugging roles using configurable model packs supporting Anthropic, OpenAI, Google, OpenRouter, Ollama, Azure, and AWS Bedrock simultaneously.
  • Full Auto Autonomous Mode
  • Executes complete development workflows autonomously including planning, implementation, command execution, and automated debugging of terminal output and browser application state without manual intervention.

Features

  • Tree-Sitter Project Maps
  • Indexes massive repositories by parsing class hierarchies, function signatures, and import graphs across 30+ languages without loading full file contents into context.
  • Interactive Terminal REPL
  • Provides fuzzy auto-complete, plan branching, step-by-step rewind, and inline diff viewing directly in the terminal for granular control over AI operations.
  • REST Orchestration API
  • Exposes 60+ endpoints for managing accounts, organizations, projects, plans, branches, context, conversation, model packs, and streaming execution programmatically.
  • Automated Debugging Loop
  • Reads terminal output and Chrome browser state to automatically identify, diagnose, and fix errors iterating until tests pass or the developer intervenes.
  • Version-Controlled Sandbox
  • Maintains branching, rewind, and selective apply capabilities on the diff sandbox enabling experimentation with alternative approaches without risking working code.