OpenAI adds custom annotation controls to the ChatGPT app browser
According to OpenAI, developers can define custom annotation controls in the ChatGPT app browser and programmatically trigger suggested prompts. Integration with WebMCP tools allows context from selected page elements to be passed to the agent.
OpenAI announced expanded annotation capabilities in the built-in browser of the ChatGPT app. According to the company, developers can define custom controls using HTML attributes or JavaScript and programmatically trigger suggested prompts. In combination with existing WebMCP tools, users can select page elements and ask their agent to take action with additional context.
The official demo shows swapping scene elements, choosing a color and filling in an expense record through a suggested prompt. The last example uses context available through the user's agent and add-ons. The announcement does not specify pricing or an exact rollout schedule.
Why it matters
The update allows prompts for the agent to be linked to a specific part of a web page. Developers can determine which controls the page offers and which prompt they trigger; the demo shows their use in editing a scene and filling in an expense record.
Two audiences, two different impacts
What this means
For individuals
When working with a page, users can select a specific element and pass it to the agent as context for a task, such as editing a scene or choosing a color.
For a business
Teams developing company websites can add custom controls for working with an agent in the ChatGPT app and connect them to existing WebMCP tools.
DevelopmentCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.