At a glance
- Availability: Experimental (how to enable).
- Auth: API key.
- Connection: The key comes from
TAVILY_API_KEY. - Docs: https://docs.tavily.com/documentation/api-reference/endpoint/search
Credentials
Set these per environment. See Connect an integration.| Variable | Required | Description |
|---|---|---|
TAVILY_API_KEY | Yes | Tavily API key (starts with tvly-) Docs. |
Setup
- Create a Tavily account: Go to https://app.tavily.com and sign up - the free plan includes monthly API credits, enough for testing.
- Copy your API key: Your default API key is shown on the dashboard home (https://app.tavily.com/home); create additional keys per project if needed.
- Store the key: Add it to your .env file as TAVILY_API_KEY=tvly-…
- Verify access: Run the Search tool with a simple query. A 401 means the key is invalid; a 432 means you exceeded your plan limit.
- All endpoints consume API credits per request; advanced search/extract depth costs more than basic
- Crawl runs synchronously and can take a while on large sites - keep max_depth and limit small
Tools
| Tool | Access | Description |
|---|---|---|
| Search | Read | Search the web and get ranked results with optional LLM answer |
| Extract | Read | Extract clean page content from one or more URLs |
| Crawl | Read | Crawl a website from a root URL and extract content from discovered pages |
| Map | Read | Get a list of URLs discovered on a website without extracting content |
Example prompts
- Search the web with Tavily for a query I provide and summarize the top results.
- Extract the content of a URL I provide with Tavily and summarize it.