agent-sdk-development
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseAgent SDK Meta Skill
Agent SDK 元技能
🚨 MANDATORY: Invoke docs-management First
🚨 强制要求:先调用docs-management
STOP - Before providing ANY response about Claude Agent SDK:
- INVOKE
skilldocs-management- QUERY using keywords: Agent SDK, TypeScript SDK, Python SDK, sessions, custom tools, permissions, or related topics
- BASE all responses EXCLUSIVELY on official documentation loaded
Skipping this step results in outdated or incorrect information.
停 - 在提供任何关于Claude Agent SDK的响应之前:
- 调用
技能docs-management- 查询时使用关键词:Agent SDK、TypeScript SDK、Python SDK、sessions、custom tools、permissions或相关主题
- 所有响应完全基于加载的官方文档
跳过此步骤会导致信息过时或错误。
Verification Checkpoint
验证检查点
Before responding, verify:
- Did I invoke docs-management skill?
- Did official documentation load?
- Is my response based EXCLUSIVELY on official docs?
If ANY checkbox is unchecked, STOP and invoke docs-management first.
在响应前,请验证:
- 我是否调用了docs-management技能?
- 官方文档是否已加载?
- 我的响应是否完全基于官方文档?
如果有任何一个复选框未勾选,请停止操作并先调用docs-management。
Overview
概述
Central authority for Claude Agent SDK (TypeScript and Python SDKs). This skill uses 100% delegation to docs-management - it contains NO duplicated official documentation.
Architecture: Pure delegation with keyword registry. All official documentation is accessed via docs-management skill queries.
Claude Agent SDK(TypeScript和Python SDK)的核心权威指南。本技能100%委托给docs-management处理,不包含任何重复的官方文档内容。
架构: 带关键词注册表的纯委托架构。所有官方文档均通过docs-management技能查询获取。
Reference Loading Guide
参考加载指南
This skill follows a pure delegation architecture with three loading layers:
本技能遵循纯委托架构,包含三个加载层级:
Layer 1: Always in Context
层级1:始终在上下文环境中
- SKILL.md (this file) - Keyword registry, decision trees, delegation patterns
- SKILL.md(本文档)- 关键词注册表、决策树、委托模式
Layer 2: On-Demand via docs-management
层级2:通过docs-management按需加载
All official Agent SDK documentation is accessed through docs-management queries:
| doc_id | Topic |
|---|---|
| platform-claude-com-docs-en-agent-sdk-overview | SDK overview and getting started |
| platform-claude-com-docs-en-agent-sdk-typescript | TypeScript SDK API reference |
| platform-claude-com-docs-en-agent-sdk-python | Python SDK API reference |
| platform-claude-com-docs-en-agent-sdk-sessions | Session management and resumption |
| platform-claude-com-docs-en-agent-sdk-streaming-vs-single-mode | Input mode selection |
| platform-claude-com-docs-en-agent-sdk-custom-tools | Creating custom tools |
| platform-claude-com-docs-en-agent-sdk-permissions | Permission configuration |
| platform-claude-com-docs-en-agent-sdk-modifying-system-prompts | System prompt customization |
| platform-claude-com-docs-en-agent-sdk-cost-tracking | Cost and usage tracking |
| platform-claude-com-docs-en-agent-sdk-todo-tracking | Todo list integration |
| platform-claude-com-docs-en-agent-sdk-structured-outputs | Structured output features |
| platform-claude-com-docs-en-agent-sdk-hosting | Production hosting patterns |
| platform-claude-com-docs-en-agent-sdk-mcp | MCP server integration |
| platform-claude-com-docs-en-agent-sdk-plugins | Plugin loading |
| platform-claude-com-docs-en-agent-sdk-subagents | Subagent usage |
| platform-claude-com-docs-en-agent-sdk-skills | Skills integration |
| platform-claude-com-docs-en-agent-sdk-slash-commands | Slash command usage |
所有官方Agent SDK文档均通过docs-management查询获取:
| doc_id | 主题 |
|---|---|
| platform-claude-com-docs-en-agent-sdk-overview | SDK概述与入门指南 |
| platform-claude-com-docs-en-agent-sdk-typescript | TypeScript SDK API参考 |
| platform-claude-com-docs-en-agent-sdk-python | Python SDK API参考 |
| platform-claude-com-docs-en-agent-sdk-sessions | 会话管理与恢复 |
| platform-claude-com-docs-en-agent-sdk-streaming-vs-single-mode | 输入模式选择 |
| platform-claude-com-docs-en-agent-sdk-custom-tools | 创建自定义工具 |
| platform-claude-com-docs-en-agent-sdk-permissions | 权限配置 |
| platform-claude-com-docs-en-agent-sdk-modifying-system-prompts | 系统提示自定义 |
| platform-claude-com-docs-en-agent-sdk-cost-tracking | 成本与使用量追踪 |
| platform-claude-com-docs-en-agent-sdk-todo-tracking | 待办事项列表集成 |
| platform-claude-com-docs-en-agent-sdk-structured-outputs | 结构化输出功能 |
| platform-claude-com-docs-en-agent-sdk-hosting | 生产环境托管模式 |
| platform-claude-com-docs-en-agent-sdk-mcp | MCP服务器集成 |
| platform-claude-com-docs-en-agent-sdk-plugins | 插件加载 |
| platform-claude-com-docs-en-agent-sdk-subagents | 子Agent使用 |
| platform-claude-com-docs-en-agent-sdk-skills | 技能集成 |
| platform-claude-com-docs-en-agent-sdk-slash-commands | 斜杠命令使用 |
Layer 3: External Resources (GitHub)
层级3:外部资源(GitHub)
- TypeScript SDK: anthropics/claude-agent-sdk-typescript
- Python SDK: anthropics/claude-agent-sdk-python
- TypeScript SDK: anthropics/claude-agent-sdk-typescript
- Python SDK: anthropics/claude-agent-sdk-python
Loading Strategy
加载策略
This skill has no local reference files. All documentation is accessed via docs-management queries using the keyword registry below. This ensures:
- Zero duplication of official documentation
- Always-current information from canonical sources
- Zero maintenance overhead for documentation sync
本技能无本地参考文件。 所有文档均通过以下关键词注册表,通过docs-management查询获取。这确保了:
- 无官方文档重复内容
- 始终获取来自权威来源的最新信息
- 无文档同步的维护开销
When to Use This Skill
何时使用本技能
Keywords: Agent SDK, Claude Agent SDK, TypeScript SDK, Python SDK, npm install claude-agent-sdk, pip install claude-agent-sdk, sessions, session resumption, forking sessions, streaming mode, single mode, custom tools, allowedTools, disallowedTools, permissionMode, system prompts, appendSystemPrompt, outputStyle, CLAUDE.md, settingSources, cost tracking, todo tracking, structured outputs, hosting, plugins SDK, MCP SDK, agent branding
Use this skill when:
- Installing or configuring Claude Agent SDK
- Setting up authentication (API key, Bedrock, Vertex)
- Managing sessions and resumption
- Choosing between streaming and single input modes
- Creating custom tools
- Configuring permissions (allowedTools, disallowedTools)
- Modifying system prompts
- Tracking costs and usage
- Working with todo lists in agents
- Getting structured outputs
- Hosting agents in production
- Integrating MCP with SDK
- Understanding SDK branding requirements
关键词: Agent SDK、Claude Agent SDK、TypeScript SDK、Python SDK、npm install claude-agent-sdk、pip install claude-agent-sdk、sessions、session resumption、forking sessions、streaming mode、single mode、custom tools、allowedTools、disallowedTools、permissionMode、system prompts、appendSystemPrompt、outputStyle、CLAUDE.md、settingSources、cost tracking、todo tracking、structured outputs、hosting、plugins SDK、MCP SDK、agent branding
在以下场景使用本技能:
- 安装或配置Claude Agent SDK
- 设置身份验证(API密钥、Bedrock、Vertex)
- 管理会话与恢复
- 选择流式输入模式与单次输入模式
- 创建自定义工具
- 配置权限(allowedTools、disallowedTools)
- 修改系统提示
- 追踪成本与使用量
- 在Agent中处理待办事项列表
- 获取结构化输出
- 在生产环境中托管Agent
- 将MCP与SDK集成
- 了解SDK品牌要求
Keyword Registry for docs-management Queries
docs-management查询用关键词注册表
Use these keywords when querying docs-management skill for official documentation:
查询docs-management技能获取官方文档时,请使用以下关键词:
SDK Fundamentals
SDK基础
| Topic | Keywords |
|---|---|
| Overview | "Agent SDK", "Claude Agent SDK overview", "why use Agent SDK" |
| Installation | "Agent SDK installation", "npm install claude-agent-sdk", "pip install claude-agent-sdk" |
| SDK Options | "SDK options", "TypeScript SDK", "Python SDK" |
| Migration | "Claude Code SDK migration", "migration guide" |
| 主题 | 关键词 |
|---|---|
| 概述 | "Agent SDK"、"Claude Agent SDK overview"、"why use Agent SDK" |
| 安装 | "Agent SDK installation"、"npm install claude-agent-sdk"、"pip install claude-agent-sdk" |
| SDK选项 | "SDK options"、"TypeScript SDK"、"Python SDK" |
| 迁移 | "Claude Code SDK migration"、"migration guide" |
Authentication
身份验证
| Topic | Keywords |
|---|---|
| API Key | "Agent SDK authentication", "ANTHROPIC_API_KEY" |
| Bedrock | "CLAUDE_CODE_USE_BEDROCK", "Bedrock SDK" |
| Vertex | "CLAUDE_CODE_USE_VERTEX", "Vertex SDK" |
| 主题 | 关键词 |
|---|---|
| API密钥 | "Agent SDK authentication"、"ANTHROPIC_API_KEY" |
| Bedrock | "CLAUDE_CODE_USE_BEDROCK"、"Bedrock SDK" |
| Vertex | "CLAUDE_CODE_USE_VERTEX"、"Vertex SDK" |
Sessions
会话
| Topic | Keywords |
|---|---|
| Session Management | "session management SDK", "session ID", "getting session ID" |
| Resumption | "session resumption", "resume option", "continuing sessions" |
| Forking | "forking sessions", "forkSession", "session branching" |
| 主题 | 关键词 |
|---|---|
| 会话管理 | "session management SDK"、"session ID"、"getting session ID" |
| 恢复 | "session resumption"、"resume option"、"continuing sessions" |
| 分叉 | "forking sessions"、"forkSession"、"session branching" |
Input Modes
输入模式
| Topic | Keywords |
|---|---|
| Overview | "streaming vs single mode", "input modes SDK" |
| Streaming Mode | "streaming input mode", "recommended mode" |
| Single Mode | "single input mode", "non-streaming" |
| 主题 | 关键词 |
|---|---|
| 概述 | "streaming vs single mode"、"input modes SDK" |
| 流式模式 | "streaming input mode"、"recommended mode" |
| 单次模式 | "single input mode"、"non-streaming" |
Custom Tools
自定义工具
| Topic | Keywords |
|---|---|
| Creating Tools | "custom tools SDK", "creating custom tools" |
| Tool Schema | "tool schema", "tool definition SDK" |
| Tool Handlers | "tool handlers", "tool execution SDK" |
| 主题 | 关键词 |
|---|---|
| 创建工具 | "custom tools SDK"、"creating custom tools" |
| 工具Schema | "tool schema"、"tool definition SDK" |
| 工具处理器 | "tool handlers"、"tool execution SDK" |
Permissions
权限
| Topic | Keywords |
|---|---|
| Tool Permissions | "allowedTools", "disallowedTools", "tool permissions SDK" |
| Permission Mode | "permissionMode SDK", "permission strategy" |
| Handling Permissions | "handling permissions SDK", "permission callbacks" |
| 主题 | 关键词 |
|---|---|
| 工具权限 | "allowedTools"、"disallowedTools"、"tool permissions SDK" |
| 权限模式 | "permissionMode SDK"、"permission strategy" |
| 权限处理 | "handling permissions SDK"、"permission callbacks" |
System Prompts
系统提示
| Topic | Keywords |
|---|---|
| Overview | "modifying system prompts", "system prompts SDK" |
| CLAUDE.md | "CLAUDE.md SDK", "settingSources", "project instructions" |
| Append System Prompt | "appendSystemPrompt", "custom instructions" |
| Output Style | "outputStyle SDK", "agent persona" |
| 主题 | 关键词 |
|---|---|
| 概述 | "modifying system prompts"、"system prompts SDK" |
| CLAUDE.md | "CLAUDE.md SDK"、"settingSources"、"project instructions" |
| 追加系统提示 | "appendSystemPrompt"、"custom instructions" |
| 输出样式 | "outputStyle SDK"、"agent persona" |
Cost and Tracking
成本与追踪
| Topic | Keywords |
|---|---|
| Cost Tracking | "cost tracking SDK", "token usage", "billing SDK" |
| Todo Tracking | "todo tracking SDK", "todo lists SDK", "task management" |
| 主题 | 关键词 |
|---|---|
| 成本追踪 | "cost tracking SDK"、"token usage"、"billing SDK" |
| 待办事项追踪 | "todo tracking SDK"、"todo lists SDK"、"task management" |
Structured Outputs
结构化输出
| Topic | Keywords |
|---|---|
| Structured Outputs | "structured outputs SDK", "JSON results", "validated outputs" |
| Schema Definition | "output schema SDK", "result validation" |
| 主题 | 关键词 |
|---|---|
| 结构化输出 | "structured outputs SDK"、"JSON results"、"validated outputs" |
| Schema定义 | "output schema SDK"、"result validation" |
Hosting and Deployment
托管与部署
| Topic | Keywords |
|---|---|
| Hosting Overview | "hosting Agent SDK", "production deployment" |
| Ephemeral Sessions | "ephemeral sessions", "pattern 1 hosting" |
| Persistent Sessions | "persistent sessions", "pattern 2 hosting" |
| 主题 | 关键词 |
|---|---|
| 托管概述 | "hosting Agent SDK"、"production deployment" |
| 临时会话 | "ephemeral sessions"、"pattern 1 hosting" |
| 持久会话 | "persistent sessions"、"pattern 2 hosting" |
Plugins and MCP
插件与MCP
| Topic | Keywords |
|---|---|
| Plugins | "plugins SDK", "loading plugins", "programmatic plugins" |
| MCP Integration | "MCP SDK", "MCP servers SDK", "extending with MCP" |
| 主题 | 关键词 |
|---|---|
| 插件 | "plugins SDK"、"loading plugins"、"programmatic plugins" |
| MCP集成 | "MCP SDK"、"MCP servers SDK"、"extending with MCP" |
Branding
品牌
| Topic | Keywords |
|---|---|
| Branding Guidelines | "agent branding", "Claude Agent naming", "SDK branding" |
| 主题 | 关键词 |
|---|---|
| 品牌指南 | "agent branding"、"Claude Agent naming"、"SDK branding" |
SDK Feature Integration
SDK功能集成
| Topic | Keywords |
|---|---|
| Subagents | "subagents SDK", "SDK subagents", "subagent integration" |
| Skills | "skills SDK", "SDK skills", "agent skills SDK" |
| Slash Commands | "slash commands SDK", "SDK commands", "custom commands SDK" |
| 主题 | 关键词 |
|---|---|
| 子Agent | "subagents SDK"、"SDK subagents"、"subagent integration" |
| 技能 | "skills SDK"、"SDK skills"、"agent skills SDK" |
| 斜杠命令 | "slash commands SDK"、"SDK commands"、"custom commands SDK" |
API Reference
API参考
| Topic | Keywords |
|---|---|
| TypeScript Reference | "TypeScript SDK reference", "TypeScript API" |
| Python Reference | "Python SDK reference", "Python API" |
| Changelog | "SDK changelog", "SDK updates" |
| 主题 | 关键词 |
|---|---|
| TypeScript参考 | "TypeScript SDK reference"、"TypeScript API" |
| Python参考 | "Python SDK reference"、"Python API" |
| 更新日志 | "SDK changelog"、"SDK updates" |
Quick Decision Tree
快速决策树
What do you want to do?
- Install the SDK -> Query docs-management: "Agent SDK installation"
- Set up authentication -> Query docs-management: "Agent SDK authentication"
- Manage sessions -> Query docs-management: "session management SDK", "resume option"
- Fork a session -> Query docs-management: "forking sessions", "forkSession"
- Choose input mode -> Query docs-management: "streaming vs single mode"
- Create custom tools -> Query docs-management: "custom tools SDK"
- Configure permissions -> Query docs-management: "allowedTools", "permissionMode SDK"
- Modify system prompts -> Query docs-management: "modifying system prompts"
- Use CLAUDE.md files -> Query docs-management: "CLAUDE.md SDK", "settingSources"
- Track costs -> Query docs-management: "cost tracking SDK"
- Add todo tracking -> Query docs-management: "todo tracking SDK"
- Get structured outputs -> Query docs-management: "structured outputs SDK"
- Host in production -> Query docs-management: "hosting Agent SDK"
- Use subagents -> Query docs-management: "subagents SDK"
- Integrate skills -> Query docs-management: "skills SDK", "agent skills SDK"
- Add slash commands -> Query docs-management: "slash commands SDK"
- Load plugins -> Query docs-management: "plugins SDK"
你想要做什么?
- 安装SDK -> 查询docs-management:"Agent SDK installation"
- 设置身份验证 -> 查询docs-management:"Agent SDK authentication"
- 管理会话 -> 查询docs-management:"session management SDK"、"resume option"
- 分叉会话 -> 查询docs-management:"forking sessions"、"forkSession"
- 选择输入模式 -> 查询docs-management:"streaming vs single mode"
- 创建自定义工具 -> 查询docs-management:"custom tools SDK"
- 配置权限 -> 查询docs-management:"allowedTools"、"permissionMode SDK"
- 修改系统提示 -> 查询docs-management:"modifying system prompts"
- 使用CLAUDE.md文件 -> 查询docs-management:"CLAUDE.md SDK"、"settingSources"
- 追踪成本 -> 查询docs-management:"cost tracking SDK"
- 添加待办事项追踪 -> 查询docs-management:"todo tracking SDK"
- 获取结构化输出 -> 查询docs-management:"structured outputs SDK"
- 生产环境托管 -> 查询docs-management:"hosting Agent SDK"
- 使用子Agent -> 查询docs-management:"subagents SDK"
- 集成技能 -> 查询docs-management:"skills SDK"、"agent skills SDK"
- 添加斜杠命令 -> 查询docs-management:"slash commands SDK"
- 加载插件 -> 查询docs-management:"plugins SDK"
Topic Coverage
主题覆盖范围
SDK Installation and Setup
SDK安装与设置
- npm package: @anthropic-ai/claude-agent-sdk
- pip package: claude-agent-sdk
- TypeScript and Python SDK options
- Migration from Claude Code SDK
- npm包:@anthropic-ai/claude-agent-sdk
- pip包:claude-agent-sdk
- TypeScript和Python SDK选项
- 从Claude Code SDK迁移
Authentication Methods
身份验证方法
- ANTHROPIC_API_KEY environment variable
- Amazon Bedrock integration (CLAUDE_CODE_USE_BEDROCK)
- Google Vertex AI integration (CLAUDE_CODE_USE_VERTEX)
- Third-party provider configuration
- ANTHROPIC_API_KEY环境变量
- Amazon Bedrock集成(CLAUDE_CODE_USE_BEDROCK)
- Google Vertex AI集成(CLAUDE_CODE_USE_VERTEX)
- 第三方提供商配置
Session Management Features
会话管理功能
- Automatic session creation
- Session ID capture from init message
- Session resumption with resume option
- Forking sessions (forkSession/fork_session)
- Session branching for experimentation
- Context preservation across sessions
- 自动会话创建
- 从初始化消息中捕获会话ID
- 使用resume选项恢复会话
- 会话分叉(forkSession/fork_session)
- 用于实验的会话分支
- 跨会话保留上下文
Input Mode Configuration
输入模式配置
- Streaming input mode (recommended)
- Single input mode
- Mode selection criteria
- Performance considerations
- 流式输入模式(推荐)
- 单次输入模式
- 模式选择标准
- 性能考量
Custom Tool Development
自定义工具开发
- Tool schema definition
- Tool handler implementation
- Tool registration with SDK
- Tool execution patterns
- 工具Schema定义
- 工具处理器实现
- 向SDK注册工具
- 工具执行模式
Permission Configuration
权限配置
- allowedTools list
- disallowedTools list
- permissionMode settings
- Permission callbacks and handling
- allowedTools列表
- disallowedTools列表
- permissionMode设置
- 权限回调与处理
System Prompt Customization
系统提示自定义
- CLAUDE.md file loading (settingSources)
- appendSystemPrompt option
- outputStyle for agent persona
- Project-level vs user-level instructions
- CLAUDE.md文件加载(settingSources)
- appendSystemPrompt选项
- 用于Agent角色的outputStyle
- 项目级与用户级指令
Cost and Usage Tracking
成本与使用量追踪
- Token usage monitoring
- Billing integration patterns
- Usage reporting
- Cost estimation
- Token使用量监控
- 账单集成模式
- 使用量报告
- 成本估算
Todo List Integration
待办事项列表集成
- Todo tracking API
- Task display and management
- Progress monitoring
- Todo state persistence
- 待办事项追踪API
- 任务显示与管理
- 进度监控
- 待办事项状态持久化
Structured Output Features
结构化输出功能
- Output schema definition
- JSON result validation
- Type-safe results
- TODO tracking agent example
- 输出Schema定义
- JSON结果验证
- 类型安全结果
- 待办事项追踪Agent示例
Hosting Patterns
托管模式
- Ephemeral sessions pattern
- Persistent sessions pattern
- Production deployment strategies
- Scaling considerations
- 临时会话模式
- 持久会话模式
- 生产环境部署策略
- 扩容考量
Full Claude Code Features
完整Claude Code功能
- Subagents (./.claude/agents/)
- Agent Skills (./.claude/skills/)
- Hooks (./.claude/settings.json)
- Slash Commands (./.claude/commands/)
- Plugins (programmatic loading)
- Memory (CLAUDE.md files)
- 子Agent(./.claude/agents/)
- Agent技能(./.claude/skills/)
- 钩子(./.claude/settings.json)
- 斜杠命令(./.claude/commands/)
- 插件(程序化加载)
- 记忆(CLAUDE.md文件)
Branding Requirements
品牌要求
- Allowed naming (Claude Agent, Claude)
- Not permitted naming (Claude Code)
- Product branding guidelines
- 允许的命名(Claude Agent、Claude)
- 不允许的命名(Claude Code)
- 产品品牌指南
Delegation Patterns
委托模式
Standard Query Pattern
标准查询模式
text
User asks: "How do I create a custom tool in the Agent SDK?"
1. Invoke docs-management skill
2. Use keywords: "custom tools SDK", "creating custom tools"
3. Load official documentation
4. Provide guidance based EXCLUSIVELY on official docstext
用户提问:“如何在Agent SDK中创建自定义工具?”
1. 调用docs-management技能
2. 使用关键词:"custom tools SDK"、"creating custom tools"
3. 加载官方文档
4. 完全基于官方文档提供指导Multi-Topic Query Pattern
多主题查询模式
text
User asks: "I want to build an agent with sessions, custom tools, and cost tracking"
1. Invoke docs-management skill with multiple queries:
- "session management SDK", "session resumption"
- "custom tools SDK", "creating custom tools"
- "cost tracking SDK", "token usage"
2. Synthesize guidance from official documentationtext
用户提问:“我想要构建一个包含会话、自定义工具和成本追踪功能的Agent”
1. 使用多个查询调用docs-management技能:
- "session management SDK"、"session resumption"
- "custom tools SDK"、"creating custom tools"
- "cost tracking SDK"、"token usage"
2. 整合官方文档中的指导内容Troubleshooting Pattern
故障排查模式
text
User reports: "My session resumption isn't working"
1. Invoke docs-management skill
2. Use keywords: "session management SDK", "resume option", "session ID"
3. Check official docs for session patterns
4. Guide user through proper session handlingtext
用户反馈:“我的会话无法恢复”
1. 调用docs-management技能
2. 使用关键词:"session management SDK"、"resume option"、"session ID"
3. 查阅官方文档中的会话模式
4. 引导用户正确处理会话Troubleshooting Quick Reference
故障排查快速参考
| Issue | Keywords for docs-management |
|---|---|
| Authentication failing | "Agent SDK authentication", "ANTHROPIC_API_KEY" |
| Session not resuming | "session resumption", "resume option" |
| Custom tool not working | "custom tools SDK", "tool handlers" |
| Permissions blocking | "allowedTools", "permissionMode SDK" |
| System prompt not loading | "CLAUDE.md SDK", "settingSources" |
| Cost tracking issues | "cost tracking SDK", "token usage" |
| Structured output invalid | "structured outputs SDK", "schema definition" |
| Streaming not working | "streaming input mode", "streaming vs single" |
| 问题 | docs-management查询关键词 |
|---|---|
| 身份验证失败 | "Agent SDK authentication"、"ANTHROPIC_API_KEY" |
| 会话无法恢复 | "session resumption"、"resume option" |
| 自定义工具无法工作 | "custom tools SDK"、"tool handlers" |
| 权限被阻止 | "allowedTools"、"permissionMode SDK" |
| 系统提示未加载 | "CLAUDE.md SDK"、"settingSources" |
| 成本追踪问题 | "cost tracking SDK"、"token usage" |
| 结构化输出无效 | "structured outputs SDK"、"schema definition" |
| 流式功能无法工作 | "streaming input mode"、"streaming vs single" |
Related Skills
相关技能
Skills that work well with agent-sdk-development:
- docs-management - Primary delegation target; provides all official Agent SDK documentation
- mcp-integration - MCP server configuration; Agent SDK can integrate with MCP servers
- settings-management - Settings configuration; SDK respects Claude Code settings
- hook-management - Hook configuration; SDK supports hooks via programmatic loading
- current-date - Get current UTC date for version history and audit timestamps
Delegation example:
text
When working with Agent SDK + MCP integration:
1. Use agent-sdk-development for SDK-specific guidance
2. Use mcp-integration for MCP server configuration
3. Both skills delegate to docs-management for official documentation与agent-sdk-development配合良好的技能:
- docs-management - 主要委托目标;提供所有官方Agent SDK文档
- mcp-integration - MCP服务器配置;Agent SDK可与MCP服务器集成
- settings-management - 配置设置;SDK遵循Claude Code设置
- hook-management - 钩子配置;SDK支持通过程序化加载钩子
- current-date - 获取当前UTC日期用于版本历史和审计时间戳
委托示例:
text
处理Agent SDK + MCP集成时:
1. 使用agent-sdk-development获取SDK相关指导
2. 使用mcp-integration获取MCP服务器配置指导
3. 两个技能均委托给docs-management获取官方文档Additional Resources
附加资源
For test scenarios, multi-model testing notes, repository-specific guidance, and external references, see:
- references/testing-and-notes.md - Test scenarios, multi-model testing, repository notes, external links
测试场景、多模型测试说明、仓库特定指导和外部参考,请查看:
- references/testing-and-notes.md - 测试场景、多模型测试、仓库说明、外部链接
Version History
版本历史
-
v1.1.0 (2025-11-27): Comprehensive audit and enhancement
- Added Reference Loading Guide with explicit doc_id table for all 17 official Agent SDK docs
- Added Test Scenarios section with 5 common use case scenarios
- Added Related Skills section documenting skill composition patterns
- Added Multi-Model Testing Notes section
- Enhanced Last Verified section with Type B audit metadata
- Verified keyword registry covers all official SDK documentation
-
v1.0.0 (2025-11-26): Initial release
- Pure delegation architecture
- Comprehensive keyword registry
- Quick decision tree
- Topic coverage for all Agent SDK features
- Troubleshooting quick reference
-
v1.1.0(2025-11-27):全面审核与增强
- 添加了参考加载指南,包含所有17份官方Agent SDK文档的明确doc_id表格
- 添加了测试场景部分,包含5个常见用例场景
- 添加了相关技能部分,记录技能组合模式
- 添加了多模型测试说明部分
- 增强了最后验证部分,包含B类审核元数据
- 验证关键词注册表覆盖所有官方SDK文档
-
v1.0.0(2025-11-26):初始版本
- 纯委托架构
- 全面的关键词注册表
- 快速决策树
- 覆盖所有Agent SDK功能的主题范围
- 故障排查快速参考
Last Updated
最后更新
Date: 2025-11-28
Model: claude-opus-4-5-20251101
日期: 2025-11-28
模型: claude-opus-4-5-20251101