HolyClaude and Anthropic’s Terms of Service

[slider23]

As is well known, Anthropic, unlike OpenAI, actively monitors the use of Claude Code subscriptions in OpenClaw and other tools and bans such accounts for violating the Terms of Service.

Quote from https://code.claude.com/docs/en/legal-and-compliance :

"Authentication and credential use

Claude Code authenticates with Anthropic’s servers using OAuth tokens or API keys. These authentication methods serve different purposes:

OAuth authentication (used with Free, Pro, and Max plans) is intended exclusively for Claude Code and Claude.ai. Using OAuth tokens obtained through Claude Free, Pro, or Max accounts in any other product, tool, or service—including the Agent SDK—is not permitted and constitutes a violation of the Consumer Terms of Service.

Developers building products or services that interact with Claude’s capabilities, including those using the Agent SDK, should use API key authentication through Claude Console or a supported cloud provider. Anthropic does not permit third-party developers to offer Claude.ai login or to route requests through Free, Pro, or Max plan credentials on behalf of their users."

Question: Does using HolyClaude violate Anthropic’s Terms of Service?

[CoderLuii]

Great question, and I want to address it directly because this is important.

HolyClaude does not violate Anthropic's Terms of Service. Here's why:

The key distinction in the docs you quoted is between using the official Claude Code CLI versus building a third-party product that routes requests through OAuth tokens.

HolyClaude runs the real, unmodified Claude Code CLI from Anthropic inside a Docker container. It does not:

• Intercept, proxy, or redirect OAuth tokens
• Route requests through Free/Pro/Max credentials on behalf of other users
• Offer Claude.ai login through a third-party interface
• Modify the CLI in any way

When you sign in through CloudCLI (the web UI), you are signing into Claude Code itself using Anthropic's own OAuth flow. The container is just the environment Claude Code runs in. It's no different from installing Claude Code on a Linux server and SSHing into it.

The ToS violation Anthropic enforces is when third-party tools extract OAuth tokens and use them in their own API calls outside of the official Claude Code client. OpenClaw and similar tools do this. HolyClaude does not. Your Claude Code session talks directly to Anthropic's servers using Anthropic's own client.

Think of it this way: running Claude Code inside Docker is like running it inside a VM or on a remote server. The container is infrastructure, not a third-party client.

That said, I'm not a lawyer and Anthropic could change their interpretation at any time. If anyone from Anthropic has concerns I'm happy to discuss. But as it stands today, this is a standard Docker deployment of their official CLI.