Token导航 LogoToken导航TokenDH.com
研究检索external-servicegithub未标认证来源可访问许可证需确认审计通过

stitch-sdk-developmentstitch SDK 开发

Agent Skill

stitch-sdk-development 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

2,117

周安装

90

GitHub Stars

1,578

下载量

742
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:stitch-sdk-development(stitch SDK 开发)
来源仓库:https://github.com/google-labs-code/stitch-sdk
仓库路径:skills/stitch-sdk-development
安装命令:
npx skills add https://github.com/google-labs-code/stitch-sdk --skill stitch-sdk-development
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/google-labs-code/stitch-sdk --skill stitch-sdk-development

简介

用于查找、检索和筛选相关信息。stitch-sdk-development 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

  • 适合根据关键词快速定位候选结果。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。
  • 可结合来源仓库和原始 README 核验具体用法。
  • 安装前建议确认权限范围和维护状态。
  • 需避免触发联网、命令执行或文件读写。

SKILL.md

Stitch SDK Development

This skill encodes the expertise needed to develop @google/stitch-sdk — the core systems, patterns, and philosophies. It does not enumerate every method (the codebase is the source of truth for that). It teaches you how to think about the system.


The Generation Pipeline

The domain layer is fully generated. No handwritten domain classes. The pipeline has 3 stages:

Stage 1: Capture            Stage 2: Domain Design       Stage 3: Generate
┌──────────────────┐       ┌──────────────────┐        ┌──────────────────────────┐
│ capture-tools.ts │──────▶│  domain-map.json │───────▶│ generate-sdk.ts          │
│                  │       │  (the IR)        │        │                          │
│ Connects to MCP  │       │ Classes, bindings│        │ Deterministic            │
│ server, calls    │       │ arg routing,     │        │ codegen into             │
│ tools/list       │       │ cache, extraction│        │ packages/sdk/generated/  │
└──────────────────┘       └──────────────────┘        └──────────────────────────┘
        │                          │                           │
        ▼                          ▼                           ▼
 tools-manifest.json        domain-map.json          packages/sdk/generated/src/*.ts
 (raw MCP tool schemas)     (tool→class mapping)     (Stitch, Project, Screen)

Stage 1 (bun scripts/capture-tools.ts): Connects to the live Stitch MCP server, calls tools/list, writes tools-manifest.json. Source of truth for what tools exist.

Stage 2 (agent/human): Reads the manifest and produces domain-map.json — the intermediate representation. This is where judgment lives: which tool maps to which class, what args come from self vs param vs computed, how to extract the return value, and what data to cache.

Stage 3 (bun scripts/generate-sdk.ts): Deterministic codegen. Reads manifest + domain-map, emits TypeScript classes in packages/sdk/generated/src/. No LLM involved — pure template expansion.

Integrity: stitch-sdk.lock records SHA-256 hashes of all inputs and outputs. bun scripts/validate-generated.ts verifies consistency. Run in CI to prevent publishing stale code.

Supporting a New Tool

When the Stitch MCP server adds a new tool:

  1. Run Stage 1 to capture the updated manifest
  2. Run Stage 2: add a binding in domain-map.json for the new tool
  3. Run Stage 3 to regenerate the SDK classes
  4. Run validate-generated.ts to confirm consistency
  5. Update tests as needed

The Domain Map IR

domain-map.json expresses two things:

Classes: What domain objects exist and how they're constructed.

{
  "Screen": {
    "constructorParams": ["projectId", "screenId"],
    "fieldMapping": {
      "projectId": { "from": "projectId" },
      "screenId": { "from": "id", "fallback": { "field": "name", "splitOn": "/screens/" } }
    },
    "parentField": "projectId",
    "idField": "screenId"
  }
}

Bindings: How MCP tools map to class methods.

{
  "tool": "generate_screen_from_text",
  "class": "Project",
  "method": "generate",
  "args": {
    "projectId": { "from": "self" },
    "prompt": { "from": "param" },
    "name": { "from": "computed", "template": "projects/{projectId}/screens/{screenId}" }
  },
  "returns": {
    "class": "Screen",
    "projection": [
      { "prop": "outputComponents", "index": 0 },
      { "prop": "design" },
      { "prop": "screens", "index": 0 }
    ]
  }
}

Arg routing: self = injected from this, param = passed by the caller, computed = built from a template at call time, selfArray = [this.field] wrapped as array.

Response projections: Structured ProjectionStep[] arrays validated against outputSchema. Use index for single items, each for arrays. Empty [] = direct return.

Cache: Methods can specify a cache with a structured projection to check this.data before calling the API:

{
  "cache": { "projection": [{ "prop": "htmlCode" }, { "prop": "downloadUrl" }], "description": "Use cached download URL from generation response" }
}

Dual Modality

The SDK serves two distinct consumers with different needs:

Agent Modality — StitchToolClient

For AI agents and orchestration scripts. Raw tool pipe. The agent receives tool schemas, constructs JSON, sends it, gets JSON back. No domain knowledge required.

Quick Start (Singleton)

The stitch singleton exposes both domain methods and tool methods via a Proxy. No instantiation needed — it lazily creates a StitchToolClient from env vars on first access.

import { stitch } from '@google/stitch-sdk';

// Discover available tools
const { tools } = await stitch.listTools();

// Call any tool by name with a JSON payload
const result = await stitch.callTool("generate_screen_from_text", {
  projectId: "123",
  prompt: "A login page",
});

// Clean up when done
await stitch.close();

The singleton reads STITCH_API_KEY (or STITCH_ACCESS_TOKEN + GOOGLE_CLOUD_PROJECT) from the environment. Set STITCH_HOST to override the server URL.

Direct Instantiation

For explicit control (multiple clients, custom config, testing), instantiate StitchToolClient directly:

import { StitchToolClient } from '@google/stitch-sdk';

const client = new StitchToolClient({ apiKey: 'my-key' });
const tools = await client.listTools();
const result = await client.callTool("create_project", { title: "My App" });
await client.close();

Config resolution: Constructor params → env vars → defaults. Auth requires either apiKey or accessToken + projectId (validated via Zod at construction time).

Connection: callTool and listTools auto-connect on first call. Concurrent calls safely share the connection via a promise-based lock.

AI SDK Adapter — stitchTools()

For agents built on the Vercel AI SDK. Transforms MCP tool schemas into AI SDK-compatible tool definitions, enabling plug-and-play with generateText().

import { stitchTools } from '@google/stitch-sdk/ai';
import { generateText } from 'ai';
import { google } from '@ai-sdk/google';

const result = await generateText({
  model: google("gemini-2.0-flash"),
  tools: stitchTools(),                          // all tools
  // or: stitchTools({ include: ["create_project"] })  // filtered
  prompt: "Create a project called My App",
});

stitchTools() is exported from the /ai subpath to keep the ai dependency optional. It uses the same shared StitchToolClient singleton internally.

stitch.toolMap provides O(1) tool lookup with pre-parsed params — static, auth-free, no network call:

const tool = stitch.toolMap.get("create_project");
tool.params;                              // ToolParam[] — flat, pre-parsed
tool.params.filter(p => p.required);      // required params only
tool.inputSchema;                         // raw ToolInputSchema still available

The raw toolDefinitions array and standalone toolMap are also exported from the main entry point.

SDK Modality — Generated Domain Classes

For humans writing precise, programmatic scripts. Generated domain facade over callTool. Typed parameters, domain objects returned, StitchError thrown on failure.

const project = await stitch.createProject("My App");
const screen = await project.generate("A login page");
const html = await screen.getHtml();

Both modalities share StitchToolClient underneath. The domain classes are a typed layer over callTool.

Error Handling — Throws at the Boundary

All generated methods use throw StitchError for error handling. No Result<T> pattern.

// Generated method pattern (inside each method):
try {
  const raw = await this.client.callTool<any>("tool_name", args);
  return /* extracted result */;
} catch (error) {
  throw StitchError.fromUnknown(error);
}

StitchError.fromUnknown() ensures all errors are normalized to StitchError with a code, message, and recoverability hint.

Infrastructure (Handwritten)

These components remain handwritten as they provide foundational plumbing:

  • StitchToolClient — MCP transport, auth, tool invocation
  • StitchError — typed error class with codes, messages, recovery hints
  • StitchProxy — MCP proxy server for re-exposing Stitch to other agents
  • singleton.ts — lazy proxy for stitch export with env var config

Traffic Light Implementation (Red → Green → Yellow)

When implementing a new feature or fixing a bug, follow the Traffic Light pattern:

🔴 Red — Write Breaking Tests

Write the test first. It must fail. This defines the contract before any implementation exists.

# Unit tests for generated classes
npx vitest run test/unit/sdk.test.ts
# → FAIL (new method doesn't exist yet)

# E2E test for the public API
bun scripts/e2e-test.ts
# → FAIL (method doesn't exist yet)

🟢 Green — Implement

  1. Add a binding to domain-map.json (Stage 2)
  2. Run bun scripts/generate-sdk.ts (Stage 3)
  3. Update tests to verify correct behavior
npx vitest run  # All tests pass
bun scripts/e2e-test.ts  # E2E passes

🟡 Yellow — Refactor / Refine / Revisit

With passing tests as your safety net:

  • Refactor for clarity (extract helpers, simplify args)
  • Add edge case tests
  • Run npx tsc to verify type safety
  • Run bun scripts/validate-generated.ts to verify pipeline integrity

Orienting in the Codebase

Discover the current state by reading the codebase directly. The key entry points:

  • Public surface: Start at packages/sdk/src/index.ts — every public export is listed here
  • Generated classes: packages/sdk/generated/src/ — Stitch, Project, Screen, DesignSystem
  • Pipeline artifacts: packages/sdk/generated/domain-map.json, packages/sdk/generated/tools-manifest.json
  • Infrastructure: packages/sdk/src/client.ts, packages/sdk/src/spec/errors.ts, packages/sdk/src/singleton.ts
  • Test structure: packages/sdk/test/unit/ for unit tests, packages/sdk/test/integration/ for live tests
  • Available commands: Read the scripts field in package.json

Do not rely on cached descriptions of files or directory trees. Read the source.

Import Convention

Use .js extensions for ESM compatibility:

import { StitchError } from '../../src/spec/errors.js';  // ✓
import { StitchError } from '../../src/spec/errors';     // ✗

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.1%
按下载量换算246

Claude

30.49%
按下载量换算226

Cursor

18.35%
按下载量换算136

Gemini CLI

8.65%
按下载量换算64

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills