websocket-googleai

Update the googleai_dart package from Google AI Live API WebSocket schema changes.

Updated May 23, 2026
One-click install
npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill websocket-googleai-kiranimmadi2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: websocket-googleai
Source: https://github.com/kiranimmadi2/promptforge-ai/tree/main/ai_clients_dart/packages/googleai_dart/.agents/skills/websocket-googleai
Command: npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill websocket-googleai-kiranimmadi2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Keeping a Dart client library in sync with the evolving Google AI Live API WebSocket schema requires manual diffing, code updates, and verification. This Skill automates the fetch, review, scaffold, and verify cycle for the googleai_dart package. ## Core Features & Use Cases - Schema Fetch & Review: Pull the latest Live API schema candidate and review changes against the canonical checked-in spec using the shared api-toolkit CLI. - Scaffolding & Verification: Scaffold new message, config, and enum types into lib/src/models/live/ and run verification checks across coverage, exports, and drift patterns. - Use Case: When Google adds a new field like proactivity to LiveConfig, run the fetch and review commands, scaffold the updated model, promote the spec into packages/googleai_dart/specs/, and verify with dart analyze and unit tests. ## Quick Start Use the websocket-googleai skill to fetch the latest Live API schema, review the changes, and update the googleai_dart package.

Frequently Asked Questions about websocket-googleai

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I update a Dart package from a WebSocket API schema change?

Run the api_toolkit fetch command with the skill's config directory to pull the latest schema, then review the diff, scaffold new types, promote the candidate spec into the package specs directory, and verify with the toolkit's verify command.

How to scaffold new Live API message types in googleai_dart?

Use the api_toolkit scaffold command with --target and --name flags against the skill config directory. The manifest.json placement rules direct generated code into lib/src/models/live/messages, config, or enums directories.

Does this workflow require Google AI credentials?

No credentials are needed for fetch, review, or verification steps since the schema is sourced from a local file. Google AI credentials are only required for actual runtime WebSocket usage of the Live API.

What verification checks run after updating the Dart models?

The toolkit verify command runs coverage and export checks against the manifest, while separate Dart quality steps run dart analyze with fatal infos, dart format, and unit tests in the googleai_dart package.

Why are some schema types marked as skipped in the manifest?

Types tagged as skipped were excluded by migration repair because their current Dart shape does not satisfy the strict shared verifier. They are acknowledged and flagged for manual model cleanup later.