Vectara Postman Collection

The Vectara Postman Collection provides a convenient way to explore and test the Vectara REST API endpoints directly in Postman. It includes pre-configured requests for common operations such as:

• Creating and managing corpora
• Indexing documents
• Querying data using Retrieval-Augmented Generation (RAG)
• Managing Vectara resources

You can fork the collection into your own Postman workspace, customize it, and start sending authenticated requests without writing code.

Prerequisites

The Vectara collection uses Postman environment variables for authentication and configuration. Ensure that you have your API key, customer ID, and corpus ID.

Variable	Description
VECTARA_API_KEY	Your API key with the required permissions
VECTARA_CUSTOMER_ID	Your Vectara customer ID
VECTARA_CORPUS_ID	The corpus identifier where your data is stored

1. Fork the Vectara Postman Collection

1. Open the Vectara Postman Collection in your browser.
2. Sign in to your Postman account.
3. Click Fork at the top of the page.
4. Select a workspace and optionally add a label or description.
5. Click Fork Collection.

The collection will now appear in your Postman app under Collections.

tip

Forking enables you to customize requests without affecting the public collection.

2. Set up authentication

Vectara supports API Key authentication for most endpoints and OAuth 2.0 for secure or automated workflows. Choose the method that fits your use case.

Option A: API Key authentication

1. In Postman, open your forked collection and select the Variables tab.
2. Add a variable named api_key and set its value to your Vectara API key (from the Vectara Console → API Keys).
3. Ensure requests use Bearer Token authentication with the token value {{api_key}}.

Option B: OAuth 2.0 authentication

1. In the collection’s Authorization tab, select OAuth 2.0.
2. Configure:
   • Grant Type: Client Credentials
   • Client ID and Client Secret from the Vectara Console
   • Access Token URL: https://api.vectara.io/oauth2/token
   • Required scopes (e.g., corpus:read, index:write)
3. Click Get New Access Token and use it for your requests.

note

For details about how to obtain credentials, see Choose the Right Authorization Method.

3. Explore and run requests

The collection contains ready-to-use requests for common API calls:

• Ask Vectara – Send a query and get a generated answer using RAG.
• Search Vectara – Perform semantic search over your corpus.
• Upload document – Add content to your corpus for retrieval and generation.
• Delete document – Remove a document from your corpus.

Each request includes the required headers and sample body content.

4. Examples

Example 1: Create a Corpus

1. Open Corpora Management, Create Corpus.
2. In the Body tab, set the corpus name and description:

   {
     "name": "product_docs",
     "description": "Product documentation corpus"
   }
3. Click Send and note the returned corpus_id.

Example 2: Ask Vectara (RAG Query)

1. Open the Querying > Ask Vectara request.
2. Update the request body:
   CREATE CORPUS EXAMPLEjson

   1

3. Click Send and review the generated answer and sources in the response.

5. Troubleshooting

• Invalid credentials: Verify your API key or OAuth token is correct and has the required permissions.
• Empty or unexpected results: Ensure you are querying the correct corpus and that it contains indexed documents.
• Connection issues: Confirm the base URL is set to https://api.vectara.io/v2/

You can view detailed request/response logs in the Postman Console (View, Show Postman Console).

Next Steps

• Explore the Vectara API Reference for detailed endpoint documentation.
• Use the collection as a foundation for building automated workflows or integrations.
• Share feedback on the collection via the Vectara Postman workspace or by contacting Vectara Support.