vercel/aiGitHub

https://github.com/vercel/ai

Stars:: 23.8k
Default branch:: main
Last scored:: 3h ago

68.0

Overall

Strengths

AGENTS.md / CLAUDE.md
CI configuration
CONTRIBUTING guide
Dependency manifest
License file

Gaps

Linter / formatter config
README

Suggestions to improve for a specific model

1
README
Expand your README to cover what the project does, how to install, the common commands, and the high-level layout.
+9.2 pts
2
Linter / formatter config
Configure a linter/formatter (ESLint+Prettier, Biome, Ruff, rustfmt+clippy, golangci-lint) and commit the config.
+7.9 pts
3
Manageable size
If possible, split into smaller modules or carve out a focused entry path. Document where to start in AGENTS.md.
+5.3 pts

Per-model scores

Claude Code

Weights AGENTS.md and tests heavily — Claude Code leans on an instructions file and a fast feedback loop.

68.9

Cursor

Weights type config and a detailed README highly — Cursor's inline edits benefit from static types and skim-readable docs.

64.4

Devin

Weights CI and reproducible envs highly — Devin runs in a sandboxed VM and needs end-to-end automation.

70.6

GPT-5 Codex

Balanced profile as a reference point.

68.2

Signal breakdown

AGENTS.md / CLAUDE.md· 100% pass

Substantive (12616 chars)

AGENTS.md

CI configuration· 100% pass

15 GitHub Actions workflow(s)

.github/workflows

CONTRIBUTING guide· 100% pass

Guide present

CONTRIBUTING.md

Dependency manifest· 100% pass

Manifest present

package.json

Reproducible dev env· 60% pass

package.json has 18 scripts

package.json

License file· 100% pass

License present

LICENSE

Linter / formatter config· 0% pass

No linter/formatter config found

Pre-commit / git hooks· 100% pass

Hook framework configured

.husky

README· 0% pass

No README found

Manageable size· 20% pass

5380+ files — very large

Test suite· 70% pass

Test files detected (packages/ai/internal/index.test.ts)

packages/ai/internal/index.test.ts

Type configuration· 100% pass

Type config present

tsconfig.json