Genkit Dart 0.16.0

Highlights

A2UI: streaming, interactive UI protocol

This release introduces A2UI (Agent-to-UI) support via the new genkit_a2ui package. An A2UI-enabled agent streams not just prose, but rich, interactive UI surfaces (components, layouts, forms) that a client (Flutter or Web) renders incrementally. The package ships the a2ui() middleware, a bundled basicCatalog, an incremental stream parser with configurable validation, and browser/Flutter-safe client helpers. A full runnable sample lives in testapps/a2ui (Shelf server plus a Flutter chat client rendering surfaces in real time). (#342, #401, #403, #404)

final uiAgent = ai.defineAgent(
  name: 'uiAgent',
  model: googleAI.gemini('gemini-flash-latest'),
  system: 'You are a helpful assistant. Render UI when a result is clearer shown than told.',
  use: [a2ui()],
);

Breaking Changes

Redesigned tool API around ToolResult and multipart (#350)

Tools now return a ToolResult<Output> instead of a bare value, using Dart dot-shorthand for ergonomics, and support multipart responses (structured output plus media parts). The stringly-typed actionType is now a typed ActionType. See the PR description for the full migration guide.

// BEFORE
fn: (input, ctx) async {
  if (ctx.resumed == null) {
    ctx.interrupt({'requiresConfirmation': true});
  }
  return 'Sunny in ${input.location}';
}

// AFTER
fn: (input, ctx) async {
  if (ctx.resumed == null) {
    return .interrupt({'requiresConfirmation': true});
  }
  return .response('Sunny in ${input.location}');
}

Key points:

• Tools are now registered and resolved under tool.v2; anything running a tool by key must use /tool.v2/<name>.
• ctx.interrupt(...) still works but is soft-deprecated.
• Reading a tool's direct result (e.g. tool.call) now yields a ToolResult: read (result as ToolResponseResult).output.
• actionType fields and params (including Registry.lookupAction) are now typed as ActionType; custom types via ActionType('my-type') still work.

New Features

• promptParts parameter across generate and generateStream (lite API, GenkitAI, and generateHelper), letting you build a user message from a list of Part objects (text and media) as an alternative to a string prompt. (#398)

final response = await ai.generate(
  promptParts: [
    TextPart(text: 'Describe this image:'),
    MediaPart(media: Media(url: 'data:image/png;base64,abc123')),
  ],
);

• Curated Gemini catalog wired into the googleAI plugin: resolve() and list() now pick up curated ModelInfo (label, stage, capabilities), with typed GoogleAiModels refs and graceful fallback to dynamic discovery. (#329)
• Curated Claude catalog for the Anthropic plugin: known Claude models resolve with typed ModelInfo, extended to 10 models with two-tier capability profiles, plus per-model thinking defaults. Unknown model names still resolve dynamically. (#322, #391, #396)

Bug Fixes

genkit (core)

• Register formatters in lite.generate so the JSON formatter's constrained default and outputInstructions are applied to lite calls. (#377)
• Map the tool role to user for Gemini API compatibility. (#340)

genkit_openai

• Stop silently dropping tools and mark o-series models as tool-capable; a model that truly lacks tools now errors loudly from the API. (#370)
• Populate usage metadata (ModelResponse.usage) for both streaming and non-streaming responses. (#373)
• Reject non-object tool input schemas with a clear INVALID_ARGUMENT error instead of a cryptic server 400. (#374)

genkit_google_genai

• Gate responseJsonSchema on constrained JSON mode and omit responseMimeType when unset. Also inherited by genkit_vertexai. (#375)
• Drop safety settings with an unset category instead of sending HARM_CATEGORY_UNSPECIFIED (which the API rejects with 400). Also inherited by genkit_vertexai. (#392)

genkit_anthropic

• Send PDF media as document blocks instead of image/png, parse the mime from the data URL, and reject malformed input locally. (#390)

New Contributors

• @adesinah made their first contribution in #396

Full Changelog: genkit-v0.15.1...genkit-v0.16.0