Automated Coding Agents Starter Kit

w/ CrewAI and Github issues

1. Limitations

1. This is not production tested
2. It is optimized for the following directory structure:

    - favicon.png
    - package.json
    - public
        - index.htm
        - favicon.png
   - src
        - App.css
        - App.js
        - index.css
      	- index.js
   
   

2. Workflow

... TODO

3. Running

1. cp env.template .env
2. python3 -m venv .venv
3. source .venv/bin/activate
4. pip3 install -r requirements.txt
5. python3 start.py to kick off the agent tasks, which will be dictated by the state of the Github repository according to Workflow above

4. Developing

1. Check out the first Demo project.
2. Its assumed that if you have an OPENROUTER_API_KEY in .env and you want to use OpenRouter.
3. Otherwise, you want to use OpenAI and OPENAI_API_KEY is required in .env
4. See .env (remember you need to create .env from env.template)

5. Vision

1. A simple, configurable human-steered A.I. web-dev GUI

pip install agents-starter-kit, Open Interpreter

6. Design Decisions

1. CrewAI is the most concise, effective agent projects we have found, and it is very wells supported by the founder & his company. So we use it instead of Autogen
2. Github is ubiquitous , reliable and fulfills all essential project-management functionality . So we use Github Issues, Pull Requests, Actions, etc.
3. OpenInterpreter should be considered (e.g. for QA Agents)

TODO:

• make "Workflow" configurable via config.yml