-
Notifications
You must be signed in to change notification settings - Fork 71
docs: Add AGENTS.md to guide AI coding assistants #320
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
docs: Add AGENTS.md to guide AI coding assistants #320
Conversation
Add structured guidance for AI assistants working with Mellea: - AGENTS.md for contributors modifying Mellea internals - docs/AGENTS_TEMPLATE.md for downstream projects to copy
|
The PR description has been updated. Please fill out the template for your PR to be reviewed. |
Merge ProtectionsYour pull request matches the following merge protections and will not be merged until they are valid. 🟢 Enforce conventional commitWonderful, this rule succeeded.Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/
|
No worries!
I haven't thought deeply about a generic Mellea AGENTS.md; I normally write a separate one for each major project I'm working on. But having a generic AGENTS definitely makes sense. I'll bring this up in our leads today. I think there are two paths:
|
- Use uv sync --all-extras --all-groups (required for tests) - Add ollama serve requirement - Remove non-existent integration marker - Fix test timing expectations (~2 min, not instant) - Remove contradictory unmarked test guidance
I'm only proposing this for mellea - it was just based on my experience with other repos, but if there is a better starting point that's cool.
It would be good to have a general feeling it's 'ok' and if there are any concerning aspects I could remove those specific statements, but we can at least start towards 2 below?
I think this is inevitable! |
Misc PR
Type of PR
Description
Adds structured guidance for AI coding assistants (Claude Code, Cursor, Copilot, etc.) working with Mellea.
Closes #319
What's included
Why
AI assistants without guidance could produce code that doesn't follow project patterns. This follows the emerging AGENTS.md convention (https://docs.anthropic.com/en/docs/claude-code/memory#agentsmd).
Feedback welcome on what to add, remove, or clarify. I missed the statement about getting comments on issue first - apologies. I'm sure the dev team will have feedback on the rules, which I built from best practice/previous projects & what I've experienced with mellea over a few days using it. I do think it's important we get something to guide towards consistency?
Testing