This add-on allows you to connect the Oxygen AI Positron plugin with AI services that expose a REST API, similar to OpenAI's chat-completion API. Unlike the built-in Open AI connector, this add-on supports the OAuth Client Credentials Flow for authentication and gives you more flexibility by letting you set query parameters.

Release Notes 1.2.0

1. Configurable AI Models - You can now manage the list of models for each AI connection from a dedicated table: add custom models, hide the models you don't want to offer, choose the recommended (default) model, and set a per-model reasoning effort level.
2. New Models Available - The GPT-5.5 Thinking, GPT-5.4, GPT-5.4 Mini, and GPT-5.4 Nano models are now available in the list of models.

Release Notes 1.1.2

1. Compatibility with the deprecated AI Positron Assistant Enterprise add-on – The add-on is now fully compatible with the AI Positron Assistant Enterprise add-on, preventing startup issues caused by remnants of the deprecated Enterprise add-on.

Release Notes 1.1.1

1. Model metadata via Positron API – Oxygen AI Positron Custom Connector now exposes detailed model information, including context window size and maximum output tokens.

Release Notes 1.1.0

1. OpenAI GPT 5 models – You can now select in the Chat view and use for invoking AI actions or for chat the latest OpenAI GPT-5, GPT-5-mini and GPT-5-nano models.
2. Connector-contributed models in Chat view – You can now browse and select models exposed by the connector directly from the Chat view.

Release Notes 1.0.2

1. Flexible OAuth2 Token Endpoint Configuration – You can now fully configure the OAuth2 token endpoint using the POSITRON_CONNECTOR_AUTH_TOKEN_URL environment variable, allowing integration with a wider range of authentication servers.
2. Support for OAuth2 Scope – Added support for specifying the OAuth2 scope via the POSITRON_CONNECTOR_AUTH_SCOPE environment variable.

Release Notes 1.0.1

1. Improved Error Handling for AI Requests – Enhanced the way errors are detected and presented when making HTTP requests to the AI service.
2. Enhanced Debug Logging – Debug logs now include additional context, making it easier for developers to troubleshoot issues.