tboulet

tboulet / Alan-Code-agent

Public

Open-source Python implementation of a Claude-Code-like agent (Gemini CLI, Codex, Copilot...). Usable in CLI, GUI, or as a python library to build upon. Works with any LiteLLM compatible API key, and several local model providers (vLLM, SGLang, Ollama).

10
0
100% credibility
Found Apr 19, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Alan Code is an open-source coding assistant that lets users chat with AI to read, edit files, run commands, and manage git in their projects via terminal or browser.

How It Works

1
🔍 Discover Alan Code

You find this helpful coding sidekick on a coding site, perfect for getting AI assistance with your projects without needing to be a tech wizard.

2
📦 Set it up quickly

Download and prepare it on your computer in moments, like installing any simple helper app.

3
🔗 Link your AI helper

Connect a smart AI service you already use, so it can understand and respond to your coding needs.

4
💬 Start chatting about code

Open it up and describe your coding problem in plain words — watch it read files, suggest fixes, and make changes right away.

5
🖥️ Pick your view

Choose the simple text chat or open a browser window for a visual overview with chat, AI's thoughts, and project history.

6
✨ Resume anytime

Pick up exactly where you left off next time, with saved conversations and smart memory of past work.

✅ Code fixed and ready

Your bugs are squashed, changes committed, and project improved — celebrate with working code and saved time!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Alan-Code-agent?

Alan-Code-agent is an open-source Python coding agent mimicking Claude Code, delivering CLI, browser GUI, or library access for tasks like file editing, bash execution, and git commits. It tackles the need for a self-hosted GitHub Copilot alternative by integrating with any LiteLLM-compatible API (Anthropic, OpenAI, Gemini) or local models via vLLM, SGLang, or Ollama. Developers get persistent sessions, cross-project memory, and live cost tracking without vendor lock-in.

Why is it gaining traction?

It stands out with unique features like a GUI showing the model's exact prompt view and a git tree for navigating agent turns, plus commands like /commit for AI-generated messages and /compact for context management. As an open-source GitHub Copilot alternative, it supports text-based tool calls for non-native models and Python APIs for auto-fix loops or custom UIs. The hook: run anywhere, from cloud APIs to local setups, with hooks and skills for extensibility.

Who should use this?

Backend devs debugging via bash/grep tools in CLI sessions, Python scripters building agent orchestrators with its library API, or indie hackers preferring self-hosted open-source GitHub tools over proprietary agents. Ideal for git-heavy workflows needing memory across sessions or cost visibility on experimental models.

Verdict

Promising early open-source Python library for agentic coding, but with just 10 stars and 1.0% credibility, it's raw—docs are solid via README/examples, yet expect bugs in git tree GUI. Try for local-model tinkering if you want a Copilot alternative; skip for production until more battle-tested.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.