Anthropic sent an takedown notice to a developer who was trying to reverse engineer its coding tool.

April 26, 2025

Image credits:Anthropic 2:57 PM PDT April 25, 2025 (19659004)

In the battle of two “agenttic” coding tool — Anthropic Claude Code and OpenAI Codex CLI — it appears that the latter is fostering more developer trust than the former. This is at least partially because Anthropic issued takedown notices against a developer who was trying to reverse-engineer Claude Code. Claude Code has a more restrictive license than Codex CLI.

Claude Code is a dueling tool that does the same thing as Codex CLI: allows developers to tap into AI models running on the cloud to complete coding tasks. Anthropic and OpenAI both released their tools within months of one another — each company racing for valuable developer mindshare.

Codex CLI source code is available under Apache 2.0 license, which allows distribution and commercial use. This is in contrast to Claude Code which is tied to Anthropic’s commercial license. This limits the ability to modify it without explicit permission from Anthropic.

Anthropic “obscured” the source code of Claude Code. Claude Code’s code is not readily available. When a Anthropic filed a lawsuit against the developer who de-obfuscated by releasing the source code to GitHub. DMCA complaint ( ) — a copyright notice requesting the removal of the code.

Developers in social media weren’t They were pleased by the move and said it compared unfavorably to OpenAI’s release of Codex CLI. OpenAI has integrated dozens of developer suggestions, including one that allows Codex CLI to be updated, into the tool’s source code in the week since Codex CLI was released. Anthropic is one of the providers that offers AI models.

Anthropic did not respond to a comment request. To be fair to Anthropic, Claude Code is in beta and a little buggy. It’s possible that Anthropic will release source code in the future under a permissive licence. Security considerations are one of the many reasons companies obfuscate code.

This is a surprising PR victory for OpenAI. In recent months, the company has resisted open source releases, preferring proprietary, locked down products. This may be a sign of a wider shift in the lab’s approach. OpenAI CEO Sam Altman said earlier this year that he thought the company was on the “wrong” side of history when it came to open source.

Kyle Wiggers, TechCrunch AI Editor. His writings have appeared in VentureBeat, Digital Trends and a variety of gadget blogs, including Android Police and Android Authority, Droid-Life and XDA-Developers. He lives in Manhattan, with his music therapist partner.

View Bio