smol developer
A junior-developer AI agent that scaffolds a whole codebase from a prompt, or can be embedded as a Python librsmol-ai/developer
- Stars
- 12.2k
- Forks
- 1.1k
- Licence
- MIT
- Language
- Python
- Last update
- 2.6 years ago
- Type
- AI model / agent
No commits in over a year — check for forks or alternatives before adopting.
What it is
smol developer is an open-source Python agent that takes a written product spec or prompt and generates a set of code files for a project. It can be run from the command line as a repo, or installed as the smol_dev library so you can call its planning and code-generation functions inside your own app. The README describes it as a way to prototype apps by refining a prompt in a loop rather than editing rigid starter templates.
- Generate a first draft of a small static web app, such as the Tic Tac Toe example in the README
- Paste an error from a generated run back into the prompt to get a revised version of the code
- Call plan, specify_file_paths, and generate_code from your own Python program
- Run the API mode via Agent Protocol to create a task and execute a step
Great for
- Developers who want to prototype a small app by iterating on a written prompt
- Developers building a feature that generates code and wants to reuse the plan, file-path, and code-generation functions
- Tinkerers curious how an LLM can break a spec into files
- People comfortable running Python and supplying their own OpenAI API key
Not ideal for
- People expecting production-ready code without review, since the README frames the output as a starting point a human must read and fix
- Non-programmers who cannot read or run the generated code
- Anyone who needs a model other than the OpenAI-based setup the README describes, since the README does not document other providers
How to get it
- 1
Clone and build from source
You'll need: Python · Poetry (install with pip install poetry if needed) · OpenAI API key
Clone the repository
git clone https://github.com/smol-ai/developer.gitMove into the folder
cd developerInstall dependencies with Poetry
poetry install - 2
Run it
- Clone the repo and install dependencies with poetry install
- Run python main.py with a short description in quotes, such as a HTML/JS/CSS Tic Tac Toe Game
- Read the generated files and identify what needs changing
- Add the fix to your prompt or prompt.md and run again
- 3
Get the most out of it
- Put longer prompts in a markdown file and use python main.py --prompt prompt.md
- Use --debug True when you need to troubleshoot a run
- Use debugger.py to ask for specific code change suggestions across the whole codebase
- Use the smol_dev library functions in your own app and add a user confirmation step after plan()
Downloads
Strengths
- Simple design: the README describes a small set of functions you can read in main.py
- Works as both a standalone command-line tool and an importable library
- Keeps a human in the loop, so you can take over the code at any point
Trade-offs
- The README relies on OpenAI models and function calling, so it likely needs an OpenAI account and API costs
- The README is truncated and examples are sparse, and its last push was in April 2024, so it may be out of date
- Generated output quality depends heavily on the prompt and the underlying model
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| developer (this) | 12.2k | AI model / agent | intermediate | 2.6 years ago | Developers who want to prototype a small app by iterating on a written prompt |
| gpt-engineer | 55k | CLI | intermediate | 1.5 years ago | Developers who want to experiment with AI-written code from a simple prompt |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.