schema-exploration

作者: langchain-ai

Lists tables, describes columns and data types, identifies foreign key relationships, and maps entity relationships in a database. Use when the user asks about…

npx skills add https://github.com/langchain-ai/deepagents --skill schema-exploration

Schema Exploration Skill

Workflow

1. List All Tables

Use sql_db_list_tables tool to see all available tables in the database.

This returns the complete list of tables you can query.

2. Get Schema for Specific Tables

Use sql_db_schema tool with table names to examine:

  • Column names - What fields are available
  • Data types - INTEGER, TEXT, DATETIME, etc.
  • Sample data - 3 example rows to understand content
  • Primary keys - Unique identifiers for rows
  • Foreign keys - Relationships to other tables

3. Map Relationships

Identify how tables connect:

  • Look for columns ending in "Id" (e.g., CustomerId, ArtistId)
  • Foreign keys link to primary keys in other tables
  • Document parent-child relationships

4. Answer the Question

Provide clear information about:

  • Available tables and their purpose
  • Column names and what they contain
  • How tables relate to each other
  • Sample data to illustrate content

Example: "What tables are available?"

Step 1: Use sql_db_list_tables

Response:

The Chinook database has 11 tables:
1. Artist - Music artists
2. Album - Music albums
3. Track - Individual songs
4. Genre - Music genres
5. MediaType - File formats (MP3, AAC, etc.)
6. Playlist - User-created playlists
7. PlaylistTrack - Tracks in playlists
8. Customer - Store customers
9. Employee - Store employees
10. Invoice - Customer purchases
11. InvoiceLine - Individual items in invoices

Example: "What columns does the Customer table have?"

Step 1: Use sql_db_schema with table name "Customer"

Response:

The Customer table has these columns:

- CustomerId (INTEGER) - Primary key, unique customer ID
- FirstName (NVARCHAR) - Customer's first name
- LastName (NVARCHAR) - Customer's last name
- Company (NVARCHAR) - Company name (optional)
- Address, City, State, Country, PostalCode - Location fields
- Phone (NVARCHAR) - Phone number
- Fax (NVARCHAR) - Fax number
- Email (NVARCHAR) - Email address
- SupportRepId (INTEGER) - Foreign key to Employee table

Sample data shows customers like:
- Luís Gonçalves from Brazil
- Leonie Köhler from Germany
- François Tremblay from Canada

Example: "How do I find revenue by artist?"

Step 1: Identify tables needed

  • Artist (has artist names)
  • Album (links artists to tracks)
  • Track (links albums to sales)
  • InvoiceLine (has sales data)
  • Invoice (has revenue totals)

Step 2: Map relationships

Artist (ArtistId)
  ↓ 1:many
Album (ArtistId, AlbumId)
  ↓ 1:many
Track (AlbumId, TrackId)
  ↓ 1:many
InvoiceLine (TrackId, UnitPrice, Quantity)

Response:

To find revenue by artist, you'll need to:
1. Start with the Artist table
2. JOIN to Album on ArtistId
3. JOIN to Track on AlbumId
4. JOIN to InvoiceLine on TrackId
5. SUM the revenue (UnitPrice * Quantity)
6. GROUP BY artist name

This requires the query-writing skill to execute.

Quality Guidelines

For "list tables" questions:

  • Show all table names
  • Add brief descriptions of what each contains
  • Group related tables (e.g., music catalog, transactions, people)

For "describe table" questions:

  • List all columns with data types
  • Explain what each column contains
  • Show sample data for context
  • Note primary and foreign keys
  • Explain relationships to other tables

For "how do I query X" questions:

  • Identify required tables
  • Map the JOIN path
  • Explain the relationship chain
  • Suggest next steps (use query-writing skill)

来自 langchain-ai 的更多技能

langgraph-docs
langchain-ai
访问LangGraph文档,构建有状态代理和多代理工作流。获取官方LangGraph Python文档,涵盖状态机、基于图的代理设计以及人机协同模式。根据查询类型优先提供相关文档:实现指南解答操作问题,概念页面讲解理论,教程提供端到端示例,API参考提供技术细节。自动选择2–4个最相关的文档URL并检索其内容以回答...
official
langgraph-human-in-the-loop
langchain-ai
暂停图执行以进行人工审查、批准或验证,随后根据其输入恢复执行。需要三个组件:检查点存储器(InMemorySaver 或 PostgresSaver)、配置中的线程 ID 以及 JSON 可序列化的中断负载。interrupt(value) 暂停执行并展示数据;Command(resume=value) 恢复执行并将该值返回给暂停的节点。恢复时,interrupt() 之前的所有代码会重新执行,因此副作用必须具有幂等性(使用 upsert 而非 insert)。支持审批工作流,...
official
web-research
langchain-ai
用于处理与网络研究相关的请求;它提供了一种结构化的方法来进行全面的网络研究
official
langchain-oss-primer
langchain-ai
任何LangChain、Deep Agents或LangGraph代理构建项目都请始终从这里开始。在选择其他技能或编写任何内容之前,这是必需的起点。
official
skill-creator
langchain-ai
创建有效技能的指南,通过专业知识、工作流程或工具集成来扩展代理能力。当用户……时使用此技能。
official
social-media
langchain-ai
根据研究内容起草特定平台的社交媒体帖子,并生成配套图片。支持领英帖子(1300字符,专业语气)和推特/X话题(每条推文280字符,采用1/🧵格式)。需在撰写前将研究任务委托给子代理,随后阅读研究结果以确保准确性和相关性。使用generate_social_image工具自动生成引人注目的社交图片,采用粗体高对比度构图,针对小屏幕进行优化...
official
deep-agents-memory
langchain-ai
为Deep Agents提供可插拔的内存与文件后端,支持临时、持久化和混合路由选项。四种后端类型:StateBackend(线程作用域,临时)、StoreBackend(跨会话持久化)、FilesystemBackend(本地开发时真实磁盘访问)和CompositeBackend(将不同路径路由到不同后端)。FilesystemMiddleware提供六种文件操作工具:ls、read_file、write_file、edit_file、glob、grep。CompositeBackend使用最长前缀匹配进行路由...
official
deep-agents-orchestration
langchain-ai
编排子代理,规划多步骤任务,并对敏感操作要求人工审批。通过任务工具将工作委派给专业子代理;自定义子代理支持独立的工具集和系统提示,而默认的“通用”子代理继承主代理配置。使用write_todos规划并跟踪复杂工作流,将任务组织为待处理、进行中和已完成状态;需要thread_id以实现跨调用的持久化。实现...
official