anyclaude CLAUDE.md

Project-specific instructions for working on anyclaude, a proxy that lets Claude Code use other AI model providers through the Anthropic API format. They describe the project, architecture, key files, and development commands.

In plain words
What is it for?
Use them when developing, debugging, or configuring anyclaude and its provider, message-conversion, streaming, or schema code.
Why use it?
They give an AI coding agent the context needed to understand how requests are translated and routed through the proxy.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/coder/anyclaude/claude-md
Clone the repo
git clone --depth 1 https://github.com/coder/anyclaude
Per session 593 This file is loaded in full into every session.
When invoked 593 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00593 $0.00593
Opus 5 $0.00296 $0.00296
Sonnet 5 $0.00119 $0.00119
Haiku 4.5 $0.00059 $0.00059

Measured 2d ago against content hash d106afb8eeaa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

anyclaude CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

anyclaude is a proxy wrapper for Claude Code that enables using alternative LLM providers (OpenAI, Google, xAI, Azure) through the Anthropic API format. It intercepts Anthropic API calls and translates them to/from the Vercel AI SDK format for the specified provider.

Architecture

The proxy works by:

  1. Spawning a local HTTP server that mimics the Anthropic API
  2. Intercepting /v1/messages requests containing <provider>/<model> format
  3. Converting Anthropic message format to AI SDK format
  4. Routing to the appropriate provider (OpenAI, Google, xAI, Azure)
  5. Converting responses back to Anthropic format
  6. Setting ANTHROPIC_BASE_URL to point Claude Code at the proxy

Key components:

  • src/main.ts: Entry point that sets up providers and spawns Claude with proxy
  • src/anthropic-proxy.ts: HTTP server that handles request/response translation
  • src/convert-anthropic-messages.ts: Bidirectional message format conversion
  • src/convert-to-anthropic-stream.ts: Stream response conversion
  • src/json-schema.ts: Schema adaptation for different providers

Development Commands

# Install dependencies
bun install

# Build the project (creates dist/main.js with shebang)
bun run build

# Run the built binary
bun run ./dist/main.js

# The build command:
# 1. Compiles TypeScript to CommonJS for Node.js compatibility
# 2. Adds Node shebang for CLI execution

Testing

Test the proxy manually:

# Run in proxy-only mode to get the URL
PROXY_ONLY=true bun run src/main.ts

# Test with a provider
OPENAI_API_KEY=your-key bun run src/main.ts --model openai/gpt-5-mini

Environment Variables

Required for each provider:

  • OPENAI_API_KEY + optional OPENAI_API_URL for OpenAI/OpenRouter
  • GOOGLE_API_KEY + optional GOOGLE_API_URL for Google
  • XAI_API_KEY + optional XAI_API_URL for xAI
  • AZURE_API_KEY + optional AZURE_API_URL for Azure
  • ANTHROPIC_API_KEY + optional ANTHROPIC_API_URL for Anthropic passthrough

Read the full file on GitHub · 73 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 73 lines · 593 tokens per session scan A d106afb8eeaa

Subscribe to this mod's changes

anyclaude CLAUDE.md is an instructions file published in the GitHub repository coder/anyclaude (292 stars, last pushed 1y ago), licensed MIT. It adds 593 tokens to every session, about $0.0030 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.