[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6991ZSB9JTM91SJ9KF7DJFP":3,"skill-same:01M6991ZSB9JTM91SJ9KF7DJFP":164},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":53,"ownerDescription":54,"updatedAt":55,"install":56,"resources":69},"01M6991ZSB9JTM91SJ9KF7DJFP","developing-genkit-dart",null,"Provides guidance on developing AI-powered applications using the Genkit Dart SDK. Use when the user asks to build AI agents in Dart, use Genkit flows, or integrate LLMs into Dart\u002FFlutter applications. Don't use for Genkit in JavaScript\u002FTypeScript, Go, or Python.","google\u002Fskills","google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","01M698VYEFDM0VJBZDYYJ2T83S","Google","01M698WR0H76ENHE587T9EJT89",23908,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"dart","genkit","developing genkit dart","flutter","go","guidance","typescript","javascript","python","user","integrate","build","when","use","flows","applications","for","ai agents","sdk","llms","2026-11-10 10:12:30",false,"---\nname: developing-genkit-dart\ndescription: >-\n  Provides guidance on developing AI-powered applications using the Genkit Dart\n  SDK. Use when the user asks to build AI agents in Dart, use Genkit flows, or\n  integrate LLMs into Dart\u002FFlutter applications. Don't use for Genkit in\n  JavaScript\u002FTypeScript, Go, or Python.\nmetadata:\n  version: \"1.0.0\"\n  category: AiAndMachineLearning\n---\n\n# Genkit Dart\n\nGenkit Dart is an AI SDK for Dart that provides a unified interface for code\ngeneration, structured outputs, tools, flows, and AI agents.\n\n## Core Features and Usage\nIf you need help with initializing Genkit (`Genkit()`), Generation\n(`ai.generate`), Tooling (`ai.defineTool`), Flows (`ai.defineFlow`), Embeddings\n(`ai.embed`), streaming, or calling remote flow endpoints, please load the core\nframework reference:\n[references\u002Fgenkit.md](references\u002Fgenkit.md)\n\n## 1.0 vs pre-1.0 APIs\n\nThis skill targets Genkit Dart **1.0** (`genkit: ^1.0.0`). Older code and\ntraining data use pre-1.0 names that no longer compile; the common ones:\n\n- HTTP: `startFlowServer` \u002F per-route `shelfHandler` \u002F `shelf_cors_headers` are\n  now `GenkitRouter` + `serve()` or `asShelfHandler()`, with `CorsOptions`;\n  context providers take `RequestData`\n  ([genkit_shelf.md](references\u002Fgenkit_shelf.md)).\n- `StatusCodes.UNAUTHENTICATED` is `StatusCode.unauthenticated` (lowerCamelCase,\n  same for `EvalStatus.pass`, `ToolChoice.required`).\n- No `RetryPlugin`; `retry()` is built in and takes `Duration`s\n  (`initialDelay`, `maxDelay`, also `cacheTtl` elsewhere).\n- Renames: `GenerateResponseHelper` -> `GenerateResult` (`modelResponse`,\n  `chunk.modelChunk`), `ExecutablePrompt` -> `Prompt\u003CInput, Output>`,\n  `embedMany` -> `embed`, `ToolFnArgs` -> `ToolFnArg`, `toolOutputSchema` ->\n  `outputSchema`, `underlyingException` -> `cause`, `modelInfo:` -> `info:`,\n  `RemoteAction.dispose()` -> `close()`,\n  `prompt(input, PromptGenerateOptions(config: ...))`\n  -> `prompt(input, config: ...)` (named options, like `generate`),\n  `defineMiddleware` ->\n  `ai.defineGenerateMiddleware` (app) \u002F `generateMiddleware` (plugins).\n- Provider config types are prefixed: `GeminiThinkingConfig`,\n  `GeminiSafetySettings`, `GeminiGoogleSearch`, `GeminiSpeechConfig`,\n  `GoogleGenAiEmbedderOptions`, `AnthropicThinkingConfig`.\n- Removed: `ToolInterruptException` \u002F `ctx.interrupt()` (return\n  `.interrupt(...)`),\n  `GoogleAiModels` \u002F `OpenAIModels` constants (use `googleAI.gemini('...')`,\n  `openAI.model('...')`). Simulated constrained output is opt-in via\n  `simulateConstrainedGeneration()`.\n- `.prompt` Picoschema is spec-compliant: `name: type, description`, never\n  `name(description): type` ([dotprompt.md](references\u002Fdotprompt.md)).\n\n## Prompts (Dotprompt)\n\n`.prompt` files keep prompt content out of Dart code with YAML frontmatter plus\na\nHandlebars template. See [references\u002Fdotprompt.md](references\u002Fdotprompt.md):\n`promptDir`, `ai.prompt()` (call\u002Fstream\u002Frender), variants, partials, named\nschemas via `defineSchema`, and the\n`tools`\u002F`maxTurns`\u002F`returnToolRequests`\u002F`use`\n(middleware) frontmatter fields. A `.prompt` file can also back an agent\ndirectly\nvia `definePromptAgent`.\n\n## Agents\n\nGenkit Dart has an **agent** API for persistent, multi-turn conversations\n(sessions, snapshots, interrupts, branching, background execution, custom state,\nartifacts, and multi-agent delegation). The agent\u002Fsession\u002Fsnapshot APIs are\n**experimental** and live behind opt-in imports: server APIs come from\n`package:genkit\u002Fexperimental.dart` (alongside `package:genkit\u002Fgenkit.dart`), the\nbrowser\u002FHTTP client from `package:genkit\u002Fexperimental_client.dart` (alongside\n`package:genkit\u002Fclient.dart`), and `dart:io` extras like `FileSessionStore` from\n`package:genkit\u002Fexperimental_io.dart`. These entry points are `@experimental`,\nso\nimporting them raises an `experimental_member_use` analyzer warning you can\nsilence in `analysis_options.yaml`. The `remoteAgent` client works from any Dart\napp, including **Flutter**, and the backend is fully interchangeable — it can\ntalk\nto a Genkit agent implemented in Dart, JS\u002FTypeScript, or Go over the same HTTP\nprotocol. A few Dart specifics: interrupts are `ai.defineInterrupt(...)` or\ntools that return `.interrupt(...)` (there is no `ctx.interrupt()`), sub-agent\ndelegation uses\nthe `agents()` middleware from `package:genkit_middleware`, and there is no\n`artifacts()` middleware yet (define artifact tools directly).\n\nFor more details see:\n\n-   [Agents](references\u002Fagents.md): defining\u002Fserving an agent and client-managed\n    state (start here).\n-   [Sessions & persistence](references\u002Fagents-sessions.md): session stores\n    (`InMemorySessionStore`\u002F`FileSessionStore`\u002F`FirestoreSessionStore`).\n-   [Human-in-the-loop \u002F interrupts](references\u002Fagents-human-in-the-loop.md):\n    pausing for approval\u002Finput via `.interrupt(...)` and resuming.\n-   [Branching](references\u002Fagents-branching.md): forking a conversation from a\n    snapshot.\n-   [Background agents](references\u002Fagents-background.md): detaching long-running\n    turns and polling.\n-   [Working with state](references\u002Fagents-state.md): typed custom session\n    state, auto-synced to the client.\n-   [Artifacts](references\u002Fagents-artifacts.md): producing and reading named\n    deliverables.\n-   [Multi-agent orchestration](references\u002Fagents-multi-agent.md): delegating to\n    sub-agents with the `agents()` middleware.\n-   [Advanced custom agents](references\u002Fagents-custom.md): `defineCustomAgent`\n    for full turn control.\n-   [Deploying agents](references\u002Fagents-deployment.md): serving agents over\n    HTTP with `GenkitRouter.addAgent` (multiple agents, shelf, CORS).\n\n## Generative UI (A2UI)\n\nGenkit Dart has an **A2UI** (Agent-to-UI) plugin (`genkit_a2ui`)\nthat lets an agent stream interactive UI **surfaces** (cards, lists, forms,\nbuttons), not just prose. The whole server-side integration is the `a2ui()`\nmodel\nmiddleware in an agent's (or `ai.generate`'s) `use` list; the Flutter client\nrenders surfaces with the [`genui`](https:\u002F\u002Fpub.dev\u002Fpackages\u002Fgenui) package plus\nthe helpers in `package:genkit_a2ui\u002Fclient.dart`. Dart specific: you must\nregister `A2uiPlugin()` in `Genkit(plugins: [...])` (unlike JS, middleware is\nresolved by name from the registry).\n\n-   [A2UI](references\u002Fa2ui.md): server middleware, options, Flutter\u002Fgenui client\n    rendering, user actions\u002Fforms, custom catalogs, and the security\u002Ftrust\n    boundary.\n\n## Genkit CLI (recommended)\n\n`genkit start` unintrusively wraps any Dart program that uses the Genkit\nlibrary, running it unchanged while capturing traces from every Genkit action so\nyou can prove tools were actually called and inspect model I\u002FO from the\nterminal, even for headless checks. It forwards stdio, so interactive CLI tools\nthat rely on stdin\u002Fstdout work without issues. Running the app directly\n(`dart run`) skips trace capture, so you're debugging blind. Check install with\n`genkit --version`.\n\n**Installation:**\n```bash\ncurl -sL cli.genkit.dev | bash # Native CLI\n# OR\nnpm install -g genkit-cli # Via npm\n# OR run commands directly with npx without a global install (prefix every genkit command):\n# npx genkit-cli start -- dart run main.dart\n```\n\n**Primary pattern (default):** prefix `genkit start --` to your normal run\ncommand. This collects telemetry from any Genkit code your program runs, whether\ntriggered from the dev UI, your own web server\u002Fweb UI, or a plain script. Starts\nthe Developer UI (usually http:\u002F\u002Flocalhost:4000) for running flows, model and\nagent playground, and browsing traces:\n\n```bash\ngenkit start -- dart run main.dart\ngenkit start --noui -- dart run main.dart   # same, without the Dev UI (still a persistent server)\n```\n`genkit start` runs until you stop it with Ctrl+C. That is expected and correct\nfor the common cases: a server your web\u002Fmobile app calls, or an interactive CLI\nyou exit yourself. `--noui` only drops the Dev UI; it is **not** a one-shot\ncommand and will not exit on its own. Do **not** use `genkit start` as a\nblocking step in automated\u002Fnon-interactive contexts; use `flow:run` (below) for\nthat.\n\n**Non-interactive use (agents\u002FCI):** add the global `--non-interactive` flag\nbefore `--` so the CLI uses defaults and never blocks on a prompt (e.g. the\nfirst-run analytics notice):\n`genkit start --non-interactive -- dart run main.dart` (works with `flow:run`\ntoo).\n\n**Run a flow (`flow:run`):** invoke a specific flow by name from the CLI. Append\nyour run command after `--` to spin up the runtime just for this run (the\ncommand runs as-is to register your flows):\n```bash\ngenkit flow:run myFlow '{\"data\": \"input\"}' -- dart run main.dart\n```\nThis is **self-terminating**: it runs the flow once, prints a `Trace ID`, then\nexits, so it's the right choice for a quick, non-interactive check (unlike\n`genkit start`). Note: `flow:run` runs **flows** (`ai.defineFlow`), not agents;\nyou can't `flow:run` an agent (`ai.defineAgent`) directly. To exercise an agent\nfrom the CLI, wrap one turn in a throwaway flow and run that (see\n[Agents](references\u002Fagents.md)). Traces for this run can be inspected using the\ntrace commands below.\n\n**Gotcha: top-level `final` declarations are lazy.** Flows and agents defined as\ntop-level `final` register with Genkit only when the symbol is first evaluated.\nAn empty `main()` registers nothing, so `flow:run` fails with\n`Process exited before runtime was ready`. Reference the flow\u002Fagent symbols from\n`main()` (or import a module that does) so their `define*` calls actually run.\n\n**Debugging with traces:** the fastest way to see prompts, model inputs\u002Foutputs,\ntool calls, latencies, and errors. Inspect from the terminal after any run under\n`genkit start`:\n```bash\ngenkit trace:list                        # find recent trace IDs\ngenkit trace:get \u003CtraceId>               # full trace details (inputs, outputs, tool calls, errors)\ngenkit trace:get \u003CtraceId> --format json # machine-readable JSON, safe to pipe into jq or other parsers\n```\n\nFor machine-readable output, pass `--format json` to get clean JSON you can pipe\ninto `jq` or other parsers. The **default** output is human-oriented (banner\u002Flog\nlines, possible truncation on large traces), so don't pipe that form directly;\nuse `--format json`, grep, or the Dev UI trace viewer.\n\n**Documentation:**\n```bash\ngenkit docs:search \"streaming\" dart\ngenkit docs:list dart\ngenkit docs:read dart\u002Fflows.md\n```\n\n## Plugin Ecosystem\nGenkit relies on a large suite of plugins to perform generative AI actions,\ninterface with external LLMs, or host web servers.\n\nWhen asked to use any given plugin, always verify usage by referring to its\ncorresponding reference below. You should load the reference when you need to\nknow the specific initialization arguments, tools, models, and usage patterns\nfor the plugin:\n\n| Plugin Name | Reference Link | Description |\n| ---- | ---- | ---- |\n| `genkit_google_genai` | [references\u002Fgenkit_google_genai.md](references\u002Fgenkit_google_genai.md) | Load for Google Gemini plugin interface usage. |\n| `genkit_vertexai` | [references\u002Fgenkit_vertexai.md](references\u002Fgenkit_vertexai.md) | Load for Gemini via Vertex AI (Google Cloud auth, versioned model IDs, embeddings incl. multimodal). |\n| `genkit_anthropic` | [references\u002Fgenkit_anthropic.md](references\u002Fgenkit_anthropic.md) | Load for Anthropic plugin interface for Claude models. |\n| `genkit_openai` | [references\u002Fgenkit_openai.md](references\u002Fgenkit_openai.md) | Load for OpenAI plugin interface for GPT models, Groq, and custom compatible endpoints. |\n| `genkit_middleware` | [references\u002Fgenkit_middleware.md](references\u002Fgenkit_middleware.md) | Load for Tooling for specific agentic behavior: `filesystem`, `skills`, and `toolApproval` interrupts. |\n| `genkit_mcp` | [references\u002Fgenkit_mcp.md](references\u002Fgenkit_mcp.md) | Load for Model Context Protocol integration (Server, Host, and Client capabilities). |\n| `genkit_chrome` | [references\u002Fgenkit_chrome.md](references\u002Fgenkit_chrome.md) | Load for Running Gemini Nano locally inside the Chrome browser using the Prompt API. |\n| `genkit_shelf` | [references\u002Fgenkit_shelf.md](references\u002Fgenkit_shelf.md) | Load for serving flows\u002Fmodels\u002Fagents over HTTP: `GenkitRouter` (core `package:genkit\u002Fio.dart`, standalone `dart:io`) and mounting it into Shelf apps. |\n| `genkit_firebase_ai` | [references\u002Fgenkit_firebase_ai.md](references\u002Fgenkit_firebase_ai.md) | Load for Firebase AI plugin interface (Gemini API via Vertex AI). |\n| `genkit_a2ui` | [references\u002Fa2ui.md](references\u002Fa2ui.md) | Load for A2UI (Agent-to-UI): streaming generative UI surfaces via the `a2ui()` middleware, rendered on the client with `genui`. |\n\n## External Dependencies\nWhenever you define schemas mapping inside of Tools, Flows, and Prompts, you\nmust use the [schemantic](https:\u002F\u002Fpub.dev\u002Fpackages\u002Fschemantic) library.\nTo learn how to use schemantic, ensure you read\n[references\u002Fschemantic.md](references\u002Fschemantic.md) for how to implement type\nsafe generated Dart code. This is particularly relevant when you encounter\nsymbols like `@Schema()`, `SchemanticType`, or classes with the `$` prefix.\nGenkit Dart uses schemantic for all of its data models so it's a CRITICAL skill\nto understand for using Genkit Dart.\n\n## Best Practices\n\n- **Agent or flow?** If the task is conversational, multi-turn, or described as\n  \"an agent\", \"assistant\", or \"chatbot\", build it with `ai.defineAgent` (see\n  [Agents](references\u002Fagents.md)) rather than hand-rolling a `generate` + tools\n  loop inside a flow. Reach for a plain flow only for single-shot, stateless\n  generation.\n- Always check that code cleanly compiles using `dart analyze` before generating\n  the final response.\n- Always use the Genkit CLI for local development and debugging.\n- Verify with traces, not a blind run. Running the app directly (`dart run`)\n  does not capture dev traces. See the [Genkit CLI](#genkit-cli-recommended)\n  section for how to run your app and capture traces.\n","# Genkit Dart\n\nGenkit Dart is an AI SDK for Dart that provides a unified interface for code\ngeneration, structured outputs, tools, flows, and AI agents.\n\n## Core Features and Usage\nIf you need help with initializing Genkit (`Genkit()`), Generation\n(`ai.generate`), Tooling (`ai.defineTool`), Flows (`ai.defineFlow`), Embeddings\n(`ai.embed`), streaming, or calling remote flow endpoints, please load the core\nframework reference:\n[references\u002Fgenkit.md](references\u002Fgenkit.md)\n\n## 1.0 vs pre-1.0 APIs\n\nThis skill targets Genkit Dart **1.0** (`genkit: ^1.0.0`). Older code and\ntraining data use pre-1.0 names that no longer compile; the common ones:\n\n- HTTP: `startFlowServer` \u002F per-route `shelfHandler` \u002F `shelf_cors_headers` are\n  now `GenkitRouter` + `serve()` or `asShelfHandler()`, with `CorsOptions`;\n  context providers take `RequestData`\n  ([genkit_shelf.md](references\u002Fgenkit_shelf.md)).\n- `StatusCodes.UNAUTHENTICATED` is `StatusCode.unauthenticated` (lowerCamelCase,\n  same for `EvalStatus.pass`, `ToolChoice.required`).\n- No `RetryPlugin`; `retry()` is built in and takes `Duration`s\n  (`initialDelay`, `maxDelay`, also `cacheTtl` elsewhere).\n- Renames: `GenerateResponseHelper` -> `GenerateResult` (`modelResponse`,\n  `chunk.modelChunk`), `ExecutablePrompt` -> `Prompt\u003CInput, Output>`,\n  `embedMany` -> `embed`, `ToolFnArgs` -> `ToolFnArg`, `toolOutputSchema` ->\n  `outputSchema`, `underlyingException` -> `cause`, `modelInfo:` -> `info:`,\n  `RemoteAction.dispose()` -> `close()`,\n  `prompt(input, PromptGenerateOptions(config: ...))`\n  -> `prompt(input, config: ...)` (named options, like `generate`),\n  `defineMiddleware` ->\n  `ai.defineGenerateMiddleware` (app) \u002F `generateMiddleware` (plugins).\n- Provider config types are prefixed: `GeminiThinkingConfig`,\n  `GeminiSafetySettings`, `GeminiGoogleSearch`, `GeminiSpeechConfig`,\n  `GoogleGenAiEmbedderOptions`, `AnthropicThinkingConfig`.\n- Removed: `ToolInterruptException` \u002F `ctx.interrupt()` (return\n  `.interrupt(...)`),\n  `GoogleAiModels` \u002F `OpenAIModels` constants (use `googleAI.gemini('...')`,\n  `openAI.model('...')`). Simulated constrained output is opt-in via\n  `simulateConstrainedGeneration()`.\n- `.prompt` Picoschema is spec-compliant: `name: type, description`, never\n  `name(description): type` ([dotprompt.md](references\u002Fdotprompt.md)).\n\n## Prompts (Dotprompt)\n\n`.prompt` files keep prompt content out of Dart code with YAML frontmatter plus\na\nHandlebars template. See [references\u002Fdotprompt.md](references\u002Fdotprompt.md):\n`promptDir`, `ai.prompt()` (call\u002Fstream\u002Frender), variants, partials, named\nschemas via `defineSchema`, and the\n`tools`\u002F`maxTurns`\u002F`returnToolRequests`\u002F`use`\n(middleware) frontmatter fields. A `.prompt` file can also back an agent\ndirectly\nvia `definePromptAgent`.\n\n## Agents\n\nGenkit Dart has an **agent** API for persistent, multi-turn conversations\n(sessions, snapshots, interrupts, branching, background execution, custom state,\nartifacts, and multi-agent delegation). The agent\u002Fsession\u002Fsnapshot APIs are\n**experimental** and live behind opt-in imports: server APIs come from\n`package:genkit\u002Fexperimental.dart` (alongside `package:genkit\u002Fgenkit.dart`), the\nbrowser\u002FHTTP client from `package:genkit\u002Fexperimental_client.dart` (alongside\n`package:genkit\u002Fclient.dart`), and `dart:io` extras like `FileSessionStore` from\n`package:genkit\u002Fexperimental_io.dart`. These entry points are `@experimental`,\nso\nimporting them raises an `experimental_member_use` analyzer warning you can\nsilence in `analysis_options.yaml`. The `remoteAgent` client works from any Dart\napp, including **Flutter**, and the backend is fully interchangeable — it can\ntalk\nto a Genkit agent implemented in Dart, JS\u002FTypeScript, or Go over the same HTTP\nprotocol. A few Dart specifics: interrupts are `ai.defineInterrupt(...)` or\ntools that return `.interrupt(...)` (there is no `ctx.interrupt()`), sub-agent\ndelegation uses\nthe `agents()` middleware from `package:genkit_middleware`, and there is no\n`artifacts()` middleware yet (define artifact tools directly).\n\nFor more details see:\n\n-   [Agents](references\u002Fagents.md): defining\u002Fserving an agent and client-managed\n    state (start here).\n-   [Sessions & persistence](references\u002Fagents-sessions.md): session stores\n    (`InMemorySessionStore`\u002F`FileSessionStore`\u002F`FirestoreSessionStore`).\n-   [Human-in-the-loop \u002F interrupts](references\u002Fagents-human-in-the-loop.md):\n    pausing for approval\u002Finput via `.interrupt(...)` and resuming.\n-   [Branching](references\u002Fagents-branching.md): forking a conversation from a\n    snapshot.\n-   [Background agents](references\u002Fagents-background.md): detaching long-running\n    turns and polling.\n-   [Working with state](references\u002Fagents-state.md): typed custom session\n    state, auto-synced to the client.\n-   [Artifacts](references\u002Fagents-artifacts.md): producing and reading named\n    deliverables.\n-   [Multi-agent orchestration](references\u002Fagents-multi-agent.md): delegating to\n    sub-agents with the `agents()` middleware.\n-   [Advanced custom agents](references\u002Fagents-custom.md): `defineCustomAgent`\n    for full turn control.\n-   [Deploying agents](references\u002Fagents-deployment.md): serving agents over\n    HTTP with `GenkitRouter.addAgent` (multiple agents, shelf, CORS).\n\n## Generative UI (A2UI)\n\nGenkit Dart has an **A2UI** (Agent-to-UI) plugin (`genkit_a2ui`)\nthat lets an agent stream interactive UI **surfaces** (cards, lists, forms,\nbuttons), not just prose. The whole server-side integration is the `a2ui()`\nmodel\nmiddleware in an agent's (or `ai.generate`'s) `use` list; the Flutter client\nrenders surfaces with the [`genui`](https:\u002F\u002Fpub.dev\u002Fpackages\u002Fgenui) package plus\nthe helpers in `package:genkit_a2ui\u002Fclient.dart`. Dart specific: you must\nregister `A2uiPlugin()` in `Genkit(plugins: [...])` (unlike JS, middleware is\nresolved by name from the registry).\n\n-   [A2UI](references\u002Fa2ui.md): server middleware, options, Flutter\u002Fgenui client\n    rendering, user actions\u002Fforms, custom catalogs, and the security\u002Ftrust\n    boundary.\n\n## Genkit CLI (recommended)\n\n`genkit start` unintrusively wraps any Dart program that uses the Genkit\nlibrary, running it unchanged while capturing traces from every Genkit action so\nyou can prove tools were actually called and inspect model I\u002FO from the\nterminal, even for headless checks. It forwards stdio, so interactive CLI tools\nthat rely on stdin\u002Fstdout work without issues. Running the app directly\n(`dart run`) skips trace capture, so you're debugging blind. Check install with\n`genkit --version`.\n\n**Installation:**\n```bash\ncurl -sL cli.genkit.dev | bash # Native CLI\n# OR\nnpm install -g genkit-cli # Via npm\n# OR run commands directly with npx without a global install (prefix every genkit command):\n# npx genkit-cli start -- dart run main.dart\n```\n\n**Primary pattern (default):** prefix `genkit start --` to your normal run\ncommand. This collects telemetry from any Genkit code your program runs, whether\ntriggered from the dev UI, your own web server\u002Fweb UI, or a plain script. Starts\nthe Developer UI (usually http:\u002F\u002Flocalhost:4000) for running flows, model and\nagent playground, and browsing traces:\n\n```bash\ngenkit start -- dart run main.dart\ngenkit start --noui -- dart run main.dart   # same, without the Dev UI (still a persistent server)\n```\n`genkit start` runs until you stop it with Ctrl+C. That is expected and correct\nfor the common cases: a server your web\u002Fmobile app calls, or an interactive CLI\nyou exit yourself. `--noui` only drops the Dev UI; it is **not** a one-shot\ncommand and will not exit on its own. Do **not** use `genkit start` as a\nblocking step in automated\u002Fnon-interactive contexts; use `flow:run` (below) for\nthat.\n\n**Non-interactive use (agents\u002FCI):** add the global `--non-interactive` flag\nbefore `--` so the CLI uses defaults and never blocks on a prompt (e.g. the\nfirst-run analytics notice):\n`genkit start --non-interactive -- dart run main.dart` (works with `flow:run`\ntoo).\n\n**Run a flow (`flow:run`):** invoke a specific flow by name from the CLI. Append\nyour run command after `--` to spin up the runtime just for this run (the\ncommand runs as-is to register your flows):\n```bash\ngenkit flow:run myFlow '{\"data\": \"input\"}' -- dart run main.dart\n```\nThis is **self-terminating**: it runs the flow once, prints a `Trace ID`, then\nexits, so it's the right choice for a quick, non-interactive check (unlike\n`genkit start`). Note: `flow:run` runs **flows** (`ai.defineFlow`), not agents;\nyou can't `flow:run` an agent (`ai.defineAgent`) directly. To exercise an agent\nfrom the CLI, wrap one turn in a throwaway flow and run that (see\n[Agents](references\u002Fagents.md)). Traces for this run can be inspected using the\ntrace commands below.\n\n**Gotcha: top-level `final` declarations are lazy.** Flows and agents defined as\ntop-level `final` register with Genkit only when the symbol is first evaluated.\nAn empty `main()` registers nothing, so `flow:run` fails with\n`Process exited before runtime was ready`. Reference the flow\u002Fagent symbols from\n`main()` (or import a module that does) so their `define*` calls actually run.\n\n**Debugging with traces:** the fastest way to see prompts, model inputs\u002Foutputs,\ntool calls, latencies, and errors. Inspect from the terminal after any run under\n`genkit start`:\n```bash\ngenkit trace:list                        # find recent trace IDs\ngenkit trace:get \u003CtraceId>               # full trace details (inputs, outputs, tool calls, errors)\ngenkit trace:get \u003CtraceId> --format json # machine-readable JSON, safe to pipe into jq or other parsers\n```\n\nFor machine-readable output, pass `--format json` to get clean JSON you can pipe\ninto `jq` or other parsers. The **default** output is human-oriented (banner\u002Flog\nlines, possible truncation on large traces), so don't pipe that form directly;\nuse `--format json`, grep, or the Dev UI trace viewer.\n\n**Documentation:**\n```bash\ngenkit docs:search \"streaming\" dart\ngenkit docs:list dart\ngenkit docs:read dart\u002Fflows.md\n```\n\n## Plugin Ecosystem\nGenkit relies on a large suite of plugins to perform generative AI actions,\ninterface with external LLMs, or host web servers.\n\nWhen asked to use any given plugin, always verify usage by referring to its\ncorresponding reference below. You should load the reference when you need to\nknow the specific initialization arguments, tools, models, and usage patterns\nfor the plugin:\n\n| Plugin Name | Reference Link | Description |\n| ---- | ---- | ---- |\n| `genkit_google_genai` | [references\u002Fgenkit_google_genai.md](references\u002Fgenkit_google_genai.md) | Load for Google Gemini plugin interface usage. |\n| `genkit_vertexai` | [references\u002Fgenkit_vertexai.md](references\u002Fgenkit_vertexai.md) | Load for Gemini via Vertex AI (Google Cloud auth, versioned model IDs, embeddings incl. multimodal). |\n| `genkit_anthropic` | [references\u002Fgenkit_anthropic.md](references\u002Fgenkit_anthropic.md) | Load for Anthropic plugin interface for Claude models. |\n| `genkit_openai` | [references\u002Fgenkit_openai.md](references\u002Fgenkit_openai.md) | Load for OpenAI plugin interface for GPT models, Groq, and custom compatible endpoints. |\n| `genkit_middleware` | [references\u002Fgenkit_middleware.md](references\u002Fgenkit_middleware.md) | Load for Tooling for specific agentic behavior: `filesystem`, `skills`, and `toolApproval` interrupts. |\n| `genkit_mcp` | [references\u002Fgenkit_mcp.md](references\u002Fgenkit_mcp.md) | Load for Model Context Protocol integration (Server, Host, and Client capabilities). |\n| `genkit_chrome` | [references\u002Fgenkit_chrome.md](references\u002Fgenkit_chrome.md) | Load for Running Gemini Nano locally inside the Chrome browser using the Prompt API. |\n| `genkit_shelf` | [references\u002Fgenkit_shelf.md](references\u002Fgenkit_shelf.md) | Load for serving flows\u002Fmodels\u002Fagents over HTTP: `GenkitRouter` (core `package:genkit\u002Fio.dart`, standalone `dart:io`) and mounting it into Shelf apps. |\n| `genkit_firebase_ai` | [references\u002Fgenkit_firebase_ai.md](references\u002Fgenkit_firebase_ai.md) | Load for Firebase AI plugin interface (Gemini API via Vertex AI). |\n| `genkit_a2ui` | [references\u002Fa2ui.md](references\u002Fa2ui.md) | Load for A2UI (Agent-to-UI): streaming generative UI surfaces via the `a2ui()` middleware, rendered on the client with `genui`. |\n\n## External Dependencies\nWhenever you define schemas mapping inside of Tools, Flows, and Prompts, you\nmust use the [schemantic](https:\u002F\u002Fpub.dev\u002Fpackages\u002Fschemantic) library.\nTo learn how to use schemantic, ensure you read\n[references\u002Fschemantic.md](references\u002Fschemantic.md) for how to implement type\nsafe generated Dart code. This is particularly relevant when you encounter\nsymbols like `@Schema()`, `SchemanticType`, or classes with the `$` prefix.\nGenkit Dart uses schemantic for all of its data models so it's a CRITICAL skill\nto understand for using Genkit Dart.\n\n## Best Practices\n\n- **Agent or flow?** If the task is conversational, multi-turn, or described as\n  \"an agent\", \"assistant\", or \"chatbot\", build it with `ai.defineAgent` (see\n  [Agents](references\u002Fagents.md)) rather than hand-rolling a `generate` + tools\n  loop inside a flow. Reach for a plain flow only for single-shot, stateless\n  generation.\n- Always check that code cleanly compiles using `dart analyze` before generating\n  the final response.\n- Always use the Genkit CLI for local development and debugging.\n- Verify with traces, not a blind run. Running the app directly (`dart run`)\n  does not capture dev traces. See the [Genkit CLI](#genkit-cli-recommended)\n  section for how to run your app and capture traces.\n",{"name":5,"metadata":44,"description":7},{"version":45,"category":46},"1.0.0","AiAndMachineLearning","main","64d010629e61e4a7d59fc21e042362f370b5888b","skills\u002Fcloud\u002Fdeveloping-genkit-dart",1994,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills","https:\u002F\u002Fgithub.com\u002Fgoogle","https:\u002F\u002Fopensource.google\u002F","Google ❤️ Open Source","2026-11-12 00:37:41",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":68},[58,61,64],{"tool":59,"command":60},"npx","npx skills add google\u002Fskills --skill developing-genkit-dart",{"tool":62,"command":63},"bunx","bunx skills add google\u002Fskills --skill developing-genkit-dart",{"tool":65,"command":66},"pnpm","pnpm dlx skills add google\u002Fskills --skill developing-genkit-dart","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002FSKILL.md",true,[70,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132,136,140,144,148,152,156,160],{"path":71,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fa2ui.md","REFERENCE","参考文档",15525,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fa2ui.md",{"path":77,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fagents-artifacts.md",6210,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-artifacts.md",{"path":81,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fagents-background.md",4178,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-background.md",{"path":85,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Fagents-branching.md",3876,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-branching.md",{"path":89,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fagents-custom.md",7079,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-custom.md",{"path":93,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Fagents-deployment.md",4456,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-deployment.md",{"path":97,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Fagents-human-in-the-loop.md",8609,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-human-in-the-loop.md",{"path":101,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":102,"sourceHtmlUrl":103},"references\u002Fagents-multi-agent.md",6188,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-multi-agent.md",{"path":105,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":106,"sourceHtmlUrl":107},"references\u002Fagents-sessions.md",7281,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-sessions.md",{"path":109,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":110,"sourceHtmlUrl":111},"references\u002Fagents-state.md",7638,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents-state.md",{"path":113,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":114,"sourceHtmlUrl":115},"references\u002Fagents.md",15691,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fagents.md",{"path":117,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":118,"sourceHtmlUrl":119},"references\u002Fdotprompt.md",8616,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fdotprompt.md",{"path":121,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":122,"sourceHtmlUrl":123},"references\u002Fgenkit_anthropic.md",1699,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_anthropic.md",{"path":125,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":126,"sourceHtmlUrl":127},"references\u002Fgenkit_chrome.md",543,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_chrome.md",{"path":129,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":130,"sourceHtmlUrl":131},"references\u002Fgenkit_firebase_ai.md",588,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_firebase_ai.md",{"path":133,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":134,"sourceHtmlUrl":135},"references\u002Fgenkit_google_genai.md",4196,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_google_genai.md",{"path":137,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":138,"sourceHtmlUrl":139},"references\u002Fgenkit_mcp.md",4650,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_mcp.md",{"path":141,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":142,"sourceHtmlUrl":143},"references\u002Fgenkit_middleware.md",5127,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_middleware.md",{"path":145,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":146,"sourceHtmlUrl":147},"references\u002Fgenkit_openai.md",2856,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_openai.md",{"path":149,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":150,"sourceHtmlUrl":151},"references\u002Fgenkit_shelf.md",4843,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_shelf.md",{"path":153,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":154,"sourceHtmlUrl":155},"references\u002Fgenkit_vertexai.md",2697,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit_vertexai.md",{"path":157,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":158,"sourceHtmlUrl":159},"references\u002Fgenkit.md",17449,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fgenkit.md",{"path":161,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":162,"sourceHtmlUrl":163},"references\u002Fschemantic.md",5358,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fdeveloping-genkit-dart\u002Freferences\u002Fschemantic.md",[165,201,231,260,287,311],{"skillNo":166,"name":167,"nameZh":6,"description":168,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":169,"categories":170,"keywords":179,"publishedAt":39},"01M6998S2EZKD3GEJ7F943RZTE","sign-in-with-google-web","Implement, configure, and secure Sign In With Google (SiwG) using Google Identity Services (GIS \u002F `https:\u002F\u002Faccounts.google.com\u002Fgsi\u002Fclient`) across web architectures. Use when creating Google sign-in buttons, implementing Google One Tap with FedCM, integrating GIS in React\u002FNext.js\u002FAngular\u002FHTML, verifying ID tokens on backend runtimes (Python\u002FNode.js\u002FGo\u002FJava), enforcing Google Workspace domain restrictions (`hd`), securing client-side ID tokens via WebCrypto nonces or encrypted IndexedDB, embedding in cross-origin iframes via the Intermediate Iframe API (`https:\u002F\u002Faccounts.google.com\u002Fgsi\u002Fintermediate`), configuring Content Security Policy (CSP), COOP, and Permissions Policy (`identity-credentials-get`) headers, or handling sign-out and revocation. Don't use for legacy Google Sign-In (`gapi.auth2`), Google Cloud IAM service accounts, or Android\u002FiOS native Credential Manager.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fidentity\u002Fsign-in-with-google-web\u002FSKILL.md",[171,176],{"categoryNo":172,"categoryName":173,"taxonomyType":174,"taxonomyTypeDesc":175},"01JBSEEDCAT000000000000009","前端开发","SCENARIO","使用场景",{"categoryNo":177,"categoryName":178,"taxonomyType":174,"taxonomyTypeDesc":175},"01JBSEEDCAT000000000000010","后端与接口",[9,180,181,182,183,184,23,185,186,187,188,189,27,190,191,192,193,194,195,196,197,198,199,200],"sign in with google web","sign","with","web","android","security","google workspace","google cloud","implement","policy","implementing","react","iam","permissions","buttons","secure","manager","identity","architectures","domain","runtimes",{"skillNo":202,"name":203,"nameZh":6,"description":204,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":205,"categories":206,"keywords":216,"publishedAt":39},"01M6998PHPDCAD684VW5WRG83D","dpop-adoption","Implement and debug OAuth 2.0 DPoP (RFC 9449) refresh token sender-constraining for WebCrypto, Node.js ES6, and browser runtimes integrating with Google's OAuth platform. Use when configuring non-extractable asymmetric key pairs (P-256), generating DPoP Proof JWTs for authorization code exchange and token refresh, or handling 400 use_dpop_nonce challenge retry loops at oauth2.googleapis.com\u002Ftoken. Don't use for unconstrained OAuth 2.0 flows (where refresh tokens are not bound to a client key pair), or for Google Cloud IAM \u002F service account authentication.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fidentity\u002Fdpop-adoption\u002FSKILL.md",[207,208,211],{"categoryNo":177,"categoryName":178,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":209,"categoryName":210,"taxonomyType":174,"taxonomyTypeDesc":175},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":212,"categoryName":213,"taxonomyType":214,"taxonomyTypeDesc":215},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[217,218,9,187,219,220,221,33,222,223,192,224,225,226,227,228,229,197,200,230,31],"dpop adoption","adoption","es6","retry","authentication","challenge","code","browser","loops","at","node","token","pair","platform",{"skillNo":232,"name":233,"nameZh":6,"description":234,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":235,"categories":236,"keywords":243,"publishedAt":39},"01M6998M72TC73EEAQH4XC7EXW","retrieving-developer-knowledge","Searches, retrieves, and synthesizes official Google developer documentation across Google Cloud, AI\u002FGemini, Android, Chrome, Web, Flutter, Go, Firebase, and other Google developer platforms. Integrates with the Developer Knowledge MCP server (search_documents, get_documents, answer_query) or the Developer Knowledge REST API fallback. Use when searching for gcloud CLI commands, API syntax, IAM permissions, official documentation, architectural comparisons, or product choice overviews. Don't use for local filesystem lookups or non-Google documentation.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Fretrieving-developer-knowledge\u002FSKILL.md",[237,238,241,242],{"categoryNo":177,"categoryName":178,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":239,"categoryName":240,"taxonomyType":174,"taxonomyTypeDesc":175},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":209,"categoryName":210,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":212,"categoryName":213,"taxonomyType":214,"taxonomyTypeDesc":215},[244,245,246,247,187,248,249,250,251,252,253,254,255,31,256,257,258,192,259,193],"retrieving developer knowledge","knowledge","developer","MCP","mcp server","official","developer documentation","get","search","local","choice","documents","query","api","across","documentation",{"skillNo":261,"name":262,"nameZh":6,"description":263,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":264,"categories":265,"keywords":266,"publishedAt":39},"01M6998J2CZEXS9BES9S3WWFKC","migrate-to-rust","Clones, analyzes, plans, and migrates legacy codebases (C++, Fortran, Java, Python, C, Shell) from a GitHub URL or local directory to a modern, high-performance, deterministic, 100% Safe Rust architecture, tracking live telemetry across every task bucket, test failure\u002Ffix cycle, and Profile-Guided Optimization (PGO) stage, and producing a first-person AI storytelling report and interactive HTML infographic. Use when migrating a GitHub repository or local codebase to Rust, porting legacy C++, Fortran, Python, or Java workloads to memory-safe Rust, designing Struct-of-Arrays (SoA) or SIMD Rust architectures, or optimizing Rust ports with LLVM PGO. Don't use for unrelated non-Rust refactoring or single-line syntax questions.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Fmigrate-to-rust\u002FSKILL.md",[],[267,268,269,270,271,272,273,274,27,275,276,277,278,279,280,253,281,282,283,31,284,285,258,286],"migrate to rust","rust","migrate","github","memory","performance","directory","safe","ports","llvm","soa","cycle","porting","fix","arrays","infographic","high-performance","URL","architecture","unrelated",{"skillNo":288,"name":289,"nameZh":6,"description":290,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":291,"categories":292,"keywords":296,"publishedAt":39},"01M6998GCC3K29WZ82KXMB9E02","finding-google-skills","Google platform decision and setup guidance, loaded on demand from Google's skill catalog. Use when a developer is choosing or setting up part of their stack, such as where to run a service, a database, storage, messaging, authentication, analytics, ads, or AI model serving, and a Google product is a reasonable candidate - whether or not a vendor is named - or when a request names a Google product or API. Brings in the matching Google skill so the answer can weigh Google options, their trade-offs, and when they are not the right fit. Skip when the stack is already settled on another provider and no Google product is named, or the task involves no platform choice.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Ffinding-google-skills\u002FSKILL.md",[293,294,295],{"categoryNo":177,"categoryName":178,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":209,"categoryName":210,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":212,"categoryName":213,"taxonomyType":214,"taxonomyTypeDesc":215},[9,297,298,299,300,230,301,302,303,304,305,221,306,307,308,254,32,309,31,310,257],"finding google skills","storage","as","messaging","model","trade-offs","provider","answer","fit","catalog","up","decision","database","candidate",{"skillNo":312,"name":313,"nameZh":6,"description":314,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":315,"categories":316,"keywords":319,"publishedAt":39},"01M6998DWNHTH0BS6W1CY8ZXMP","workload-manager-basics","Use this skill to manage Google Cloud Workload Manager evaluations, rules, scanned resources, and validation results by using public client libraries and the REST API. Use when you need to inspect workload best-practice rules, create and run evaluations for Google Cloud general best practices, SAP, SQL Server, or custom organizational rules, review violations, export results to BigQuery, or automate Workload Manager through client libraries because no service-specific public CLI or MCP server is available. Don't use for general Google Compute Engine instance management, VPC configuration, or standard IAM auditing.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fworkload-manager-basics\u002FSKILL.md",[317,318],{"categoryNo":177,"categoryName":178,"taxonomyType":174,"taxonomyTypeDesc":175},{"categoryNo":209,"categoryName":210,"taxonomyType":174,"taxonomyTypeDesc":175},[320,321,196,322,9,247,323,187,248,324,325,326,327,31,328,257,192,329,330,331,332],"workload manager basics","workload","basics","cli","review","vpc","resources","inspect","you","create","libraries","server","run"]