Creating a text agent with web search

Get API authentication credentials as described in Authentication with the Yandex AI Studio API.

1. Create a search index.

2. Create a file named index.py and add the following code to it:

   import openai
   import json
   
   YANDEX_CLOUD_FOLDER = "..."
   YANDEX_CLOUD_API_KEY = "..."
   YANDEX_CLOUD_MODEL = "yandexgpt"
   
   client = openai.OpenAI(
       api_key=YANDEX_CLOUD_API_KEY,
       base_url="https://rest-assistant.api.cloud.yandex.net/v1",
       project=YANDEX_CLOUD_FOLDER
   )
   
   response = client.responses.create(
       model=f"gpt://{YANDEX_CLOUD_FOLDER}/{YANDEX_CLOUD_MODEL}",
       input="Give me a quick overview of LLM news for September 2025.",
       tools=[
           {
               "web_search": {
                   "filters": {
                       "allowed_domains": [
                           "habr.ru"
                       ]
                   },
                   "user_location": {
                       "region": "213",
                   }
               }
           }
       ],
       temperature=0.3,
       max_output_tokens=1000
   )
   
   # Response text
   print("Response text:")
   print(response.output_text)
   print("\n" + "=" * 50 + "\n")
   
   # Full response
   print("Full response (JSON):")
   print(json.dumps(response.model_dump(), indent=2, ensure_ascii=False))xs
   

3. Save authentication data into environment variables:

   export YANDEX_CLOUD_FOLDER=<folder_ID>
   export YANDEX_CLOUD_API_KEY=<API_key>
   export VECTOR_STORE_ID=<search_index_ID>
   

4. Run the file you created:

   python index.py