Agentforce Data Library Basics — 代理数据库基础（练习操作步骤）

本文是《Agentforce Data Library Basics — 代理数据库基础》的练习操作步骤。阅读原文请见：Agentforce Data Library Basics — 代理数据库基础。

1. Create a Data Library

1. From Setup, search for and select Agentforce Data Library.
2. From the Agentforce Data Library setup page, click New Library.
3. Enter a unique library name (e.g. Coral Cloud Spa Services) and an optional description reflecting its purpose or content.
4. Click Save.

2. Choose a Data Source

1. Choose the type of data source: Knowledge articles or file uploads. (Once chosen, you can't change it later.)
2. For file-based: Click Upload Files and select the files you want to include.
   • Text/HTML files: max 4 MB each
   • PDF files: max 100 MB each
3. Click Save.
   • Behind the scenes: data streams are pushed to Data 360, data objects are created and mapped, and a search index and retriever are created automatically.

3. Assign Data Libraries to an Agent

1. Click the name of your agent, then click Open in Builder.
2. Click the Knowledge tab.
3. Select a library for your agent (or click New Library to create one).
4. Save your changes.
5. Click Activate.

4. Test the Data Library in Preview

1. Ensure the search index and data library are ready.
2. In Agentforce Builder, open the Preview panel. Restart the conversation if needed.
3. Ask the agent a question that requires information from the uploaded files.
   • Example: What spa services are available for sensitive skin?
4. Review the agent's response — it should cite information from the files in the data library.
5. Ask follow-up questions to verify details:
   • Example: Are there any seasonal specials for sensitive skin?
   • Example: What's the price of the Holiday Bliss Package?
6. Expand the agent's reasoning to verify it's citing files from your data library.
7. Click Activate to deploy the agent with the data library.