OpenAI

🔑 BYO Key — AI

AI-powered text and image generation. Use GPT for content, translations, and summaries. Use DALL-E 3 or GPT Image for AI-generated images. Bring your own API key.

🔑 Setup Required — Bring Your Own Key This integration requires API credentials. Store them once via setup_integration or the Dashboard Vault — they're encrypted with AES-256-GCM and never exposed.

Required Credentials

openai_api_key

API Key

Find your API key at platform.openai.com → API keys. Starts with "sk-". Usage is billed to your OpenAI account.

Get your API key at platform.openai.com →

Try it with your AI assistant

Once WebsitePublisher is connected to your AI assistant, paste one of these into the chat. Replace the parts in [brackets] with your own details.

ChatGPT Claude Other assistants

@WebsitePublisher add a chat assistant to my site that answers visitor questions, powered by my own OpenAI key

@WebsitePublisher use my OpenAI account to write product descriptions for my [10] newest products

@WebsitePublisher translate my [page] into [Spanish] with OpenAI and publish it as a new page

With WebsitePublisher, add a chat assistant to my site that answers visitor questions, powered by my own OpenAI key

With WebsitePublisher, use my OpenAI account to write product descriptions for my [10] newest products

With WebsitePublisher, translate my [page] into [Spanish] with OpenAI and publish it as a new page

With WebsitePublisher, add a chat assistant to my site that answers visitor questions, powered by my own OpenAI key

With WebsitePublisher, use my OpenAI account to write product descriptions for my [10] newest products

With WebsitePublisher, translate my [page] into [Spanish] with OpenAI and publish it as a new page

Endpoints (4)

POST chat-completion

Generate text using GPT. Use for content writing, translations, summaries, extraction, and classification. Returns the model response.

Field	Type	Required	Description
`model`	string	✓ Yes	Model id, required. Any chat model your API key can use, e.g. "gpt-4.1-mini". Call list-models for the full list; OpenAI retires older names.
`messages`	array	✓ Yes	Array of messages: [{role: "system"\|"user"\|"assistant", content: "..."}]. System message sets behavior, user message is the prompt.
`max_completion_tokens`	integer	No	Maximum tokens in the response, reasoning tokens included. Reasoning models (o-series, gpt-5 family) accept only this limit, not max_tokens. Leave out for OpenAI's default.
`max_tokens`	integer	No	Legacy response limit for older models (gpt-4o, gpt-4.1). Reasoning models reject it; use max_completion_tokens there. 1 token ≈ 0.75 words.
`temperature`	number	No	Creativity 0 to 2 (default 1). Lower = more focused, higher = more creative. Reasoning models may only accept the default; leave it out for those.
`response_format`	object	No	Optional output format, passed to OpenAI unchanged. {"type": "json_object"} for valid JSON, or {"type": "json_schema", "json_schema": {"name": "verdict", "schema": {...}, "strict": true}} to force the reply to match a JSON Schema. With json_object, the messages must also ask for JSON. Leave out for plain text.

POST generate-image

Generate an image from a text prompt using DALL-E 3 or GPT Image. Returns the image URL (valid for 1 hour) or base64 data.

Field	Type	Required	Description
`prompt`	string	✓ Yes	Text description of the image to generate. Be specific and detailed for best results.
`model`	string	No	Model: "dall-e-3" (default, high quality) or "gpt-image-1" (newer, more creative).
`size`	string	No	Image size. DALL-E 3: "1024x1024" (default), "1024x1792" (portrait), "1792x1024" (landscape). GPT Image: also supports "256x256", "512x512".
`quality`	string	No	"standard" (default, faster) or "hd" (more detailed, DALL-E 3 only).
`style`	string	No	"vivid" (default, dramatic) or "natural" (more realistic, DALL-E 3 only).
`n`	integer	No	Number of images to generate (1-4 for GPT Image, always 1 for DALL-E 3).
`response_format`	string	No	"url" (default, temporary URL valid ~1h) or "b64_json" (base64 data, use for permanent storage via upload_asset).

GET list-models

List all available models on your OpenAI account. Use to verify which models you have access to.

No input parameters required.

POST create-embedding

Generate text embeddings for semantic search, similarity matching, or classification. Returns a vector of floats.

Field	Type	Required	Description
`input`	string	✓ Yes	Text to embed. Max ~8000 tokens for text-embedding-3-small.
`model`	string	✓ Yes	Embedding model id, required, e.g. "text-embedding-3-small" or "text-embedding-3-large" (more dimensions). Use the same model for everything you compare.

MCP Tool Names

When using this integration through an AI assistant (Claude, ChatGPT, Cursor, etc.), the endpoints are available as MCP tools:

Endpoint	MCP Tool Name
chat-completion	`openai_chat_completion`
generate-image	`openai_generate_image`
list-models	`openai_list_models`
create-embedding	`openai_create_embedding`

← Back to all integrations