Embed Agent Friendly Code Score Badge

[hsnice16]

Hi team — I'm Himanshu, I built Agent Friendly Code, which scores public repos on how legible they are to AI coding agents (clear conventions, docs, tests, build signals — not anything about accepting agent-authored PRs).

appwrite scored 77.8/100 — full breakdown: https://www.agentfriendlycode.com/repo/41

If you're open to it, here's a badge you can drop in the README:

[![Agent Friendly](https://agentfriendlycode.com/api/badge/github/appwrite/appwrite.svg)](https://agentfriendlycode.com/repo/41)

Renders as:

A note on what this isn't: the badge signals codebase readability for agents, not an invitation for drive-by AI PRs. I just wanted to let you know that your contribution policy is unchanged. Totally fine to pass — happy either way, and feedback on the score itself is welcome.

---

[hardikkaurani]

This is actually an interesting direction, especially as AI coding agents become more integrated into day-to-day development workflows.

A score focused on “agent readability” rather than “AI-generated contribution acceptance” is an important distinction. Things like:

• clear project structure
• strong documentation
• deterministic build steps
• discoverable conventions
• reliable tests
• typed APIs
• explicit contribution patterns

benefit both humans and AI systems trying to understand large repositories.

In practice, repositories that are easier for agents to reason about are usually also easier for:

• new contributors onboarding
• code search/navigation
• automated tooling
• static analysis
• documentation generation
• maintenance over time

The important nuance is exactly what you mentioned: this does not imply maintainers want low-quality drive-by AI PRs. Repository readability and contribution policy are separate concerns.

It would actually be interesting to see more detail about how the scoring is weighted. For example:

• documentation coverage
• architectural discoverability
• test reliability
• dependency clarity
• typed interfaces
• CI reproducibility
• API surface consistency
• monorepo navigation
• naming conventions
• code modularity

A transparent scoring rubric would probably make the metric more actionable for maintainers looking to improve their repositories intentionally.

One potentially useful future direction could be:

• historical trend tracking
• repo-to-repo comparisons by category
• suggestions for improving weak scoring areas
• CI integration checks
• “agent onboarding” recommendations generated automatically

As AI-assisted development grows, “machine-readable maintainability” will probably become a real software quality dimension alongside traditional DX and maintainability metrics.

Related references:

• [Appwrite Repository](https://github.com/appwrite/appwrite?utm_source=chatgpt.com)
• [Software Maintainability Overview](https://en.wikipedia.org/wiki/Maintainability?utm_source=chatgpt.com)
• [CI/CD Best Practices](https://docs.github.com/en/actions/guides/about-continuous-integration?utm_source=chatgpt.com)

Interesting project overall — especially because it focuses on repository structure/readability rather than simply measuring AI-generated contribution volume.

[krishkhinchi]

Interesting idea honestly. A repo being “agent friendly” is mostly about good structure, docs, and consistency, which also helps human contributors a lot.

Would be cool to see a transparent breakdown of how the score is calculated though.

[hsnice16]

I think you can go through these two pages.

Breakdown of the Appwrite - https://www.agentfriendlycode.com/repo/41
Methodology - https://www.agentfriendlycode.com/methodology