Skip to content

README.en.md

Latest commit

 

History

History
128 lines (92 loc) · 3.57 KB

README.en.md

File metadata and controls

128 lines (92 loc) · 3.57 KB
Raw

VEGA Chat API Skill

English | Русский

A skill for AI agents (Cursor, Claude Code, Windsurf, etc.) that enables building applications, services, and integrations using the VEGA Chat API — an OpenRouter proxy at api.vega.chat.

What is this

VEGA Chat API is a full OpenRouter proxy. The entire OpenRouter feature set is supported — the only difference is the base URL:

https://api.vega.chat/api/v1

Any code written for OpenRouter works with VEGA by simply replacing https://openrouter.ai/api/v1 with https://api.vega.chat/api/v1.

Features

  • Chat Completions — streaming and non-streaming, tool calling, structured outputs
  • Models — list models, filtering, get by slug
  • Embeddings — text and image vector embeddings
  • Responses API — OpenAI Responses API
  • Anthropic Messages — Anthropic Messages API
  • Audio — text-to-speech (TTS) and speech-to-text (STT)
  • Images — image generation and analysis
  • Video Generation — text-to-video and image-to-video
  • Rerank — document reranking by relevance
  • Generation Metadata — generation stats, token counts, costs
  • Error Handling — error codes, streaming errors, retry logic

Installation

Via Skills CLI

npx skills add lingeniare/api-vega-chat -g -y

Manually

Copy the api-vega-chat/ folder into your AI agent's skills directory:

  • Cursor: ~/.cursor/skills/
  • Claude Code: ~/.claude/skills/
  • Windsurf: ~/.windsurf/skills/

Usage

Once installed, the skill automatically activates when you ask your AI agent to build an application using VEGA API, api.vega.chat, or any VEGA endpoint.

Example prompt

Build a Python chatbot that streams responses from the openai/gpt-4o model using VEGA Chat API

Python quick start

from openai import OpenAI

client = OpenAI(
    base_url="https://api.vega.chat/api/v1",
    api_key="YOUR_VEGA_API_KEY",
)

response = client.chat.completions.create(
    model="openai/gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}],
)

print(response.choices[0].message.content)

TypeScript quick start

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.vega.chat/api/v1',
  apiKey: 'YOUR_VEGA_API_KEY',
});

const completion = await client.chat.completions.create({
  model: 'openai/gpt-4o',
  messages: [{ role: 'user', content: 'Hello!' }],
});

console.log(completion.choices[0].message);

Getting an API key

  1. Go to https://api.vega.chat/keys
  2. Create a key, set a name and optionally a credit limit
  3. Use the key in the Authorization: Bearer YOUR_VEGA_API_KEY header

Structure

api-vega-chat/
├── SKILL.md                          # Main skill file
├── references/                       # Detailed endpoint documentation
│   ├── chat-completions.md
│   ├── models.md
│   ├── embeddings.md
│   ├── responses-api.md
│   ├── anthropic-messages.md
│   ├── audio.md
│   ├── images.md
│   ├── video-generation.md
│   ├── rerank.md
│   ├── generation-metadata.md
│   └── error-handling.md
└── evals/
    └── evals.json                    # Test prompts

Documentation

License

MIT