from trending-skills42
Routes Claude Code tool calls through an OpenAI-compatible shim so Claude Code can use GPT-4o, Gemini, Ollama and other OpenAI-style models while preserving too
OpenClaude provides an OpenAI-compatible shim for Claude Code, translating Anthropic-style tool and stream interfaces into OpenAI Chat Completions calls. This lets Claude Code continue to use its full tool system (bash, file ops, webfetch, agents, notebooks) while running on a wide range of models: GPT-4o, DeepSeek, Gemini (via OpenRouter), Ollama, Groq, Mistral, Azure OpenAI, and local LM hosts.
Use this skill when you need Claude Code's agent/tooling surface but prefer or must run on a non-Anthropic provider (OpenAI, Ollama, OpenRouter-backed Gemini, or other OpenAI-compatible endpoints). It's useful for teams migrating from Anthropic, testing alternative models, or consolidating tooling around OpenAI-compatible APIs.
Primarily targets Claude Code (Anthropic-based agent stacks) but enables compatibility with any agent runtime that can call OpenAI-style chat completions (OpenAI, OpenRouter-backed Gemini, Ollama local runtimes, Codex).
OpenClaude Multi-LLM is a skill that routes Claude Code tool calls through an OpenAI-compatible shim, enabling use of GPT-4o, DeepSeek, Gemini, Ollama and other models. No bundled scripts were present for testing. The SKILL.md is thorough with clear provider configuration examples, troubleshooting, and architecture documentation. Minor security note: installs a scoped npm package globally and references a non-standard git clone URL (node.gitlawb.com).
No scripts to run. Skill is documentation-only for a third-party fork of Claude Code. The fork itself (openclaude) would need separate security review. The SKILL.md is well-written and comprehensive. The unusual git clone domain (node.gitlawb.com) is worth noting but not inherently malicious.
