HomeProjectsArticlesPK BoardAbout
Explore Projects

DeepCode

Active
GitHub Python MIT

Description

DeepCode is an open agentic coding platform supporting Paper2Code, Text2Web, and Text2Backend, leveraging agent technology for automated software development workflows.

Key Features

  • Multi-agent coding system with separate Planning, Implementation, and Review agents for structured development
  • Paper2Code capability to convert research papers into working code implementations
  • Text2Web and Text2Backend for generating web frontends and backend services from natural language descriptions
  • Dual interface with CLI for terminal-based development and Web UI for visual interactive experience
  • OpenRouter model selector for choosing from a wide catalog of LLM providers
  • Research-backed architecture with published arXiv paper validating multi-agent code generation approach

Use Cases

💡 Converting academic research papers into reproducible code implementations
💡 Rapid prototyping of full-stack web applications from text descriptions
💡 Automated backend service generation for API development tasks
💡 Code review and quality assurance through multi-agent collaboration
💡 Learning and education: understanding how agents decompose and implement coding tasks

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (16.5k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

coding python llm agent automation

Categories

💻 Coding Agent

Quick Start

Clone the repo, install Python 3.13+, configure your LLM API keys in the settings, choose between CLI (`python -m deepcode`) or Web UI mode, provide a task description or paper URL, and let the multi-agent system plan, implement, and review the code.
Visit GitHub

Related Projects

AutoCodeRover

3.1k · Python
Stale C

AutoCodeRover is a project structure-aware autonomous software engineer agent that achieves automated program repair and issue resolution by understanding the overall codebase architecture.

codingpythonagent +2
  • · Fully autonomous bug fixing and feature addition on GitHub issues using LLM-powered program analysis
  • · Program Structure Aware code search via abstract syntax tree analysis for precise context retrieval
  • · Statistical fault localization using test cases to achieve higher repair rates when tests are available
View on GitHub

Mini SWE-Agent

6.9k · Python
Active A+

Mini SWE-Agent is a minimalist AI agent in just 100 lines of code that solves GitHub issues or assists developers in the command line, demonstrating core coding agent capabilities with minimal implementation.

codingpythonagent +2
  • · Minimal ~100-line Python agent class achieving >74% on SWE-bench verified
  • · Bash-only tool interface requiring no special LLM tool-calling capabilities
  • · Completely linear message history for easy debugging and fine-tuning
View full details → View on GitHub

Kodus AI

1.3k · TypeScript
Active B

Kodus AI is an open-source AI code review tool with full control over model choice and costs, automatically analyzing pull requests and delivering high-quality code review feedback.

codingtypescriptagent +2
  • · Model agnostic — supports Claude, GPT-5, Gemini, Llama, GLM, Kimi and any OpenAI-compatible endpoint
  • · Zero markup on LLM costs — pay model providers directly with no hidden multipliers
  • · Kody agent learns your code architecture, standards and workflows for personalized reviews
View on GitHub