GemCode is a terminal-based AI coding agent that uses Google's Gemma models via NVIDIA NIM API to plan, write, execute, test, and review code in a loop with a live-streaming TUI.

Stars

6

7-day growth

No data

Forks

0

Open issues

0

License

MIT

Last updated

2026-07-30

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It combines a planner agent with an automated Writer→Executor→Tester→Reviewer loop, supports custom skill injection, and provides real-time streaming output, making it a practical tool for rapid Python prototyping and learning agentic workflows.

Who it is for

  • Python developers looking to automate code generation and testing
  • AI/ML enthusiasts exploring agentic coding with Gemma models
  • Hackathon participants needing rapid prototyping help
  • Terminal lovers who prefer CLI-based AI tools

Use cases

  • Prototyping small Python applications from a natural language goal
  • Automating the cycle of writing, executing, and testing code with feedback
  • Learning how multi-agent coding loops work in practice
  • Injecting domain-specific guidelines via custom skills for tailored outputs

Strengths

  • Uses Google’s Gemma models through NVIDIA NIM API, accessible with a free key
  • Rich TUI with live streaming of thinking time, token usage, and output
  • Custom skills allow injecting external knowledge into the Planner agent
  • Robust loop with retries and feedback until tests pass or limit reached

Considerations

  • Tasks execute sequentially; no parallel execution of independent tasks
  • Generated code runs locally in a temporary subprocess without container sandboxing
  • Optimized for Python only; other languages require extending executor.py
  • Dependence on NVIDIA NIM API entitlements and rate limits for specific Gemma models

README quick start

Installation & Setup

Related repositories

Similar projects matched by category, topics, and programming language.

jamesob
Featured
jamesob GitHub avatar

local-llm

A comprehensive guide for building and configuring a high-end local machine to run state-of-the-art LLMs, with detailed hardware choices, BIOS tuning, and Docker-based model serving.

AI & Machine LearningLarge Language Models
1,660
0xwilliamortiz
Featured
0xwilliamortiz GitHub avatar

openclaude-improved

OpenClaude is an open-source CLI coding agent that runs on any platform and supports a wide range of LLM providers, offering the same tools and workflows as Claude Code.

AI & Machine LearningLarge Language Models
577
mereyabdenbekuly-ctrl
Featured
mereyabdenbekuly-ctrl GitHub avatar

clodex-ide

Clodex is an open-source, local-first agentic IDE that combines persistent AI tasks, code, terminal, browser, Git, models, memory, and governed execution in one Electron workspace, currently in technical preview.

AI & Machine LearningAI Agents
859