Skip to Content

Connect Neon AI Gateway to Your AI Agent

Call AI models through a Neon branch gateway

7 toolsAPI key

20,000 free app calls every month(paid services billed separately)

Try a Neon AI Gateway Task

Copy a prompt into your agent with the oo Skill installed. Add your page links and source notes before running it.

oo SkillNeon AI Gateway

Call AI models through a Neon branch gateway

Call AI models through a Neon branch gateway

/oo Neon AI Gateway: List the models available on my Neon branch, then summarize [text] using [model].

Start Working with Neon AI Gateway

  1. Install OOMOL

    Set up oo CLI and the oo Skill in your agent.

    Install OOMOL →
  2. Connect Neon AI Gateway

    Connect a branch gateway token.

    Connect Neon AI Gateway→
  3. Give Your Agent a Task

    Use /oo with the pages to read, the changes to make, and where to save the result.

    Connection Guide →
  • A Neon branch credential with ai_gateway:invoke.
  • The branch gateway URL from the Neon Console.
  • Streaming inference is not supported by these actions.

Neon AI Gateway Tools

Call AI models through a Neon branch gateway

Call AI models through a Neon branch gateway

Call AI models through a Neon branch gateway

7 tools

Create Chat Completion

Create a chat completion through Neon AI Gateway's OpenAI-compatible `/v1/chat/completions` endpoint. One branch credential reaches Anthropic, OpenAI, Google, and open-weight models.

neon_ai_gateway.create_chat_completion

Create Response

Create a response through Neon AI Gateway's OpenAI Responses API endpoint (`/openai/v1/responses`). Required for `gpt-5-3-codex` and `gpt-5-5-pro`. Image generation needs streaming, which connector actions do not support, because non-streaming responses are capped at 655,360 bytes.

neon_ai_gateway.create_response

Create Message

Create an Anthropic Messages API message through Neon AI Gateway (`/anthropic/v1/messages`). Use it for prompt caching and extended thinking on Claude models.

neon_ai_gateway.create_message

Create Embedding

Create embeddings through Neon AI Gateway's OpenAI-compatible `/v1/embeddings` endpoint.

neon_ai_gateway.create_embedding

Generate Content

Generate content with a Google Gemini model through Neon AI Gateway's native Gemini endpoint (`/gemini/v1beta/models/{model}:generateContent`). Use it for Gemini-only options such as `thinkingConfig`; the chat completions action also serves Gemini models.

neon_ai_gateway.generate_content

Evaluate

Answer yes/no (noul), choice, and score questions about text, JSON, or images with TypeSafe System One decision models through Neon AI Gateway (`/typesafe/v1/systemone`). Returns probabilities instead of generated text.

neon_ai_gateway.evaluate

List Models

List the Neon AI Gateway model catalog in an OpenRouter-shaped response. Each entry's `enabled` field says whether this account can call the model now.

neon_ai_gateway.list_models