Fuck Claude

About

Who We Are

LinXiaoTao — Independent developer who built Fuck Claude from public reverse-engineering of Claude Code fingerprinting. This site is a reference tool and educational knowledge base — not affiliated with Anthropic.

What Fuck Claude Is

Fuck Claude is an open-source technical diagnostic tool and educational knowledge base. It was developed to help software engineers, researchers, and global developers analyze how modern AI client software (specifically Anthropic's Claude Code and Claude web interfaces) inspects system environments and performs locale fingerprinting.

Through ten distinct browser-accessible signals — including system timezone, language order, Chinese core fonts, hardware vendor fonts, WebRTC candidate leaks, domestic browser engines, device vendor attributes, Intl locale resolution, UTC offset, and emoji glyph metrics — our open scanner provides a transparent, weighted 0–100 risk score and clear debugging recommendations.

Editorial Standards & Research Methodology

All content published on Fuck Claude adheres to rigorous software engineering and technical verification standards:

  • Empirical Reverse-Engineering: Our guides and scoring weights are grounded in reproducible reverse-engineering analyses of publicly distributed binaries, prompt injection diagnostics, and open-source network captures.
  • Continuous Validation: When Anthropic publishes updates, modifies model deprecation schedules, adjusts API rate limits, or updates terms of service, our team verifies and updates the corresponding Anti-Ban Guides and Claude News archive.
  • Peer Review & Open Source Transparency: The entire website, detection algorithms, and guide repositories are open source under the MIT License on GitHub. Community feedback and pull requests undergo technical review before merging.

Technical Capabilities & Scope Limitations

Our browser scanner measures indicators that a client-side web application can legitimately observe. However, it represents an educational simulation rather than Anthropic's proprietary backend risk infrastructure:

  • Anthropic's complete risk engine incorporates server-side signals such as IP autonomous system number (ASN) classification, datacenter vs residential reputation scores, payment-card Bank Identification Number (BIN) velocity, account lifecycle metadata, and token consumption patterns.
  • A clean 0/100 scan score confirms that your client-side browser and OS environment exhibit no obvious leakage tells; it is not an absolute warranty against platform rate limits or administrative account reviews.

Open Source & Community Collaboration

We believe in open knowledge and software transparency. The project source code is available at LinXiaoTao/FuckClaude. For technical bug reports, algorithm suggestions, or editorial corrections, visit our contact page or submit an issue on GitHub.

Disclaimer of Affiliation

Fuck Claude is an independent open-source educational project. It is not affiliated with, maintained by, authorized by, or endorsed by Anthropic, PBC. Claude, Claude Code, Anthropic, and associated trademarks belong to Anthropic, PBC.