DuckAI2API — AI skill for Claude Code
OpenAI / Anthropic / Responses API server for Duck.ai — free GPT-5.6 Luna reasoning, real SSE streaming, Claude Code tool-loop.
How to install DuckAI2API
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open czpGhost/DuckAI2API and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What DuckAI2API does
OpenAI / Anthropic / Responses API server for Duck.ai — free GPT-5.6 Luna reasoning, real SSE streaming, Claude Code tool-loop. Cross-platform.
Alternatives in AI
- Deepreasoning — A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthr 5.4k ★
- Cumora — Where agent teams gather 3.2k ★
- WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★
README
🦆 DuckAI2API
**Turn Duck.ai into an OpenAI / Anthropic / Responses compatible API — for free.**
[](LICENSE) [](https://github.com/czpGhost/DuckAI2API/stargazers) [](https://github.com/czpGhost/DuckAI2API/network/members) [](https://www.python.org) [](#requirements) [](#models)
[🇨🇳 中文](./README.zh.md) · [🇬🇧 English](./README.md)
A relay that speaks **three** LLM protocols at once, backed by **Duck.ai** (`https://duck.ai/`) — no API key, no payment. Built from a js-reverse analysis of Duck.ai's live web protocol (XHR + SSE, no official API), re-verified after every major site change.
Works as a drop-in backend for **Claude Code**, the **OpenAI SDK**, and the **OpenAI Responses API**. All chat is streamed straight from Duck.ai's real SSE feed.
✨ Why
Duck.ai has **no public API**. Its chat runs over:
POST https://duck.ai/duckchat/api/status— fetchserver_vqd_02(X-Vqd-4 token)POST https://duck.ai/duckchat/api/refs— fetchvqd(X-Vqd-4 token for chat)POST https://duck.ai/duckchat/api/requests/chatwithAccept: text/event-stream— SSE response
The model id is sent as `model: ""` in the JSON body. The relay maps friendly model names to these ids.
🚀 Features
| Capability | |
|---|---|
| 🧠 9 free models | GPT-5.6 Luna, GPT-5.4 (+mini), Claude Sonnet/Haiku/Opus 4.x, Mistral Small, gpt-oss 120B, Gemma 4 31B |
| 🔌 Triple protocol | /v1/chat/completions (OpenAI), /v1/messages (Anthropic), /v1/responses (OpenAI Responses) |
| 🌊 *Real streaming |
Related Skills
Yaak Plugin SSE Parser
A Yaak plugin that parses Server-Sent Events (SSE) streaming responses from LLM APIs into readable text, with
Gptcc
GPT for Claude Code — use OpenAI GPT models inside the Claude Code CLI via documented extension points (ANTHRO
My Free Code
Open-source multi-provider AI gateway for Claude Code and other coding agents, with model routing, streaming,
Bring Your Own Model
Run Claude Code on any OpenAI or OpenAI-compatible model — a tiny, dependency-free local bridge that translate
Nova Agent
AI agent on Android — 27 native phone tools (camera, GPS, SMS, sensors, files), 3 AI providers (OpenAI · Anthr
Claude Skill GPT Pro
Claude Code skill for cross-model review. One command packages files + prompt for ChatGPT Pro deep reasoning.
Related Agents
LLM Integrator
LLM integration specialist who connects to OpenAI/Anthropic/Ollama APIs, designs prompt templates, implements
GPT Agent
GPT-5.4/o4 tasks via AI Gateway — alternative perspective, advanced reasoning, OpenAI ecosystem
GPT
Use this agent when you need to leverage GPT-5's superior capabilities for complex analysis, deep reasoning, o