Skip to content
Agent Friendly Code

Claude Code vs Codex: which do top repos set up for?

Claude Code averages higher: 71.9 for Claude Code and 70.5 for Codex, across 402 repos. Claude Code scores higher on 287, Codex on 107, and 8 tie.

Scores measure how well each repo is set up for an agent: instruction files, tests, CI and the other checks, weighted by what each agent's docs say it uses, not which agent writes better code.

Which files does each one read?

Reads from the repo
Claude Code: CLAUDE.md.claude/CLAUDE.md.claude/rules/CLAUDE.local.mdAGENTS.md.claude/AGENTS.md
Codex: AGENTS.mdAGENTS.override.md.agents/skills/
Reads AGENTS.md
Claude Code: Only when there is no CLAUDE.md
Codex: Always
Reads CLAUDE.md
Claude Code: Yes
Codex: No
From its docs
Claude Code: Reads AGENTS.md only when no CLAUDE.md, .claude/CLAUDE.md or CLAUDE.local.md is found (v2.1.277 and later). A CLAUDE.md that imports @AGENTS.md loads both, as does a setting.
Codex: Reads AGENTS.md before any work, one per directory from the repo root down to the working directory, up to 32 KiB combined; AGENTS.override.md replaces the AGENTS.md beside it.
Repos with its own file
Claude Code: CLAUDE.md only: 18 of 402 (4%)
Codex: AGENTS.md: 207 of 402 (51%)

AGENTS.md and CLAUDE.md mean the file at the repo root; a repo with both counts under AGENTS.md, and empty files don't count. Every agent's files.

Which checks does each weigh more?

Share of each agent's score that a check carries, where Claude Code and Codex differ most
CheckClaude CodeCodex
Skills (.agents/skills)0%5%
Test suite13%10%
Reproducible dev env12%9%
CI configuration7%9%
AGENTS.md / CLAUDE.md13%11%
Type configuration8%9%

Each figure is the check's share of the agent's total weight, from the docs linked below. A check few repos pass, such as an agent's own file, lowers that agent's scores wherever it carries weight. The methodology lists every weight.

Which repos are set up better for Claude Code than Codex?

Which repos are set up better for Codex than Claude Code?

Which one leads in each language?

Average Claude Code and Codex scores of the repos in each language; the higher is in bold
LanguageReposClaude CodeCodex
TypeScript8377.075.7
Python5675.674.1
Rust4574.974.0
Go4076.774.6
Java2866.965.5
C++2164.062.4
JavaScript1579.277.1

What if I use both?

  1. Write the shared instructions once, in AGENTS.md at the repo root. What to put in it.
  2. Claude Code: import @AGENTS.md in a CLAUDE.md; without that, a CLAUDE.md hides AGENTS.md from it.
  3. Codex: reads AGENTS.md before any work, up to 32 KiB combined across directories.

AGENTS.md vs CLAUDE.md · Which files each agent reads · Make a repo agent-ready

From the docs: code.claude.com/memory (new tab) · code.claude.com/skills (new tab) · learn.chatgpt.com/agents-md (new tab) · learn.chatgpt.com/build-skills (new tab)

Boards: best set up for Claude Code · best set up for Codex

More comparisons: Claude Code vs Cursor · Gemini CLI vs Claude Code · Cursor vs Codex · Aider vs Claude Code · OpenHands vs Claude Code · Pi vs Claude Code · Kimi Code CLI vs Claude Code · Devin vs Claude Code · Codex vs Gemini CLI