Tools
Tavily
Search and extract information
Integrate Tavily into the workflow. Can search the web and extract content from specific URLs. Requires API Key.
Perform AI-powered web searches using Tavily's search API. Returns structured results with titles, URLs, snippets, and optional raw content, optimized for relevance and accuracy.
| Parameter | Type | Required | Description |
|---|
query | string | Yes | The search query to execute |
max_results | number | No | Maximum number of results (1-20) |
topic | string | No | Category type: general, news, or finance (default: general) |
search_depth | string | No | Search scope: basic (1 credit) or advanced (2 credits) (default: basic) |
include_answer | string | No | LLM-generated response: true/basic for quick answer or advanced for detailed |
include_raw_content | string | No | Parsed HTML content: true/markdown or text format |
include_images | boolean | No | Include image search results |
include_image_descriptions | boolean | No | Add descriptive text for images |
include_favicon | boolean | No | Include favicon URLs |
chunks_per_source | number | No | Maximum number of relevant chunks per source (1-3, default: 3) |
time_range | string | No | Filter by recency: day/d, week/w, month/m, year/y |
start_date | string | No | Earliest publication date (YYYY-MM-DD format) |
end_date | string | No | Latest publication date (YYYY-MM-DD format) |
include_domains | string | No | Comma-separated list of domains to whitelist (max 300) |
exclude_domains | string | No | Comma-separated list of domains to blacklist (max 150) |
country | string | No | Boost results from specified country (general topic only) |
auto_parameters | boolean | No | Automatic parameter configuration based on query intent |
apiKey | string | Yes | Tavily API Key |
| Parameter | Type | Description |
|---|
query | string | The search query that was executed |
results | array | Search results with titles, URLs, content snippets, and optional metadata |
answer | string | LLM-generated answer to the query (if requested) |
images | array | Query-related images (if requested) |
auto_parameters | object | Automatically selected parameters based on query intent (if enabled) |
response_time | number | Time taken for the search request in seconds |
Extract raw content from multiple web pages simultaneously using Tavily's extraction API. Supports basic and advanced extraction depths with detailed error reporting for failed URLs.
| Parameter | Type | Required | Description |
|---|
urls | string | Yes | URL or array of URLs to extract content from |
extract_depth | string | No | The depth of extraction (basic=1 credit/5 URLs, advanced=2 credits/5 URLs) |
format | string | No | Output format: markdown or text (default: markdown) |
include_images | boolean | No | Incorporate images in extraction output |
include_favicon | boolean | No | Add favicon URL for each result |
apiKey | string | Yes | Tavily API Key |
| Parameter | Type | Description |
|---|
results | array | Successfully extracted content from URLs |
failed_results | array | URLs that failed to extract content |
response_time | number | Time taken for the extraction request in seconds |
Systematically crawl and extract content from websites using Tavily's crawl API. Supports depth control, path filtering, domain restrictions, and natural language instructions for targeted crawling.
| Parameter | Type | Required | Description |
|---|
url | string | Yes | The root URL to begin the crawl |
instructions | string | No | Natural language directions for the crawler (costs 2 credits per 10 pages) |
max_depth | number | No | How far from base URL to explore (1-5, default: 1) |
max_breadth | number | No | Links followed per page level (≥1, default: 20) |
limit | number | No | Total links processed before stopping (≥1, default: 50) |
select_paths | string | No | Comma-separated regex patterns to include specific URL paths (e.g., /docs/.*) |
select_domains | string | No | Comma-separated regex patterns to restrict crawling to certain domains |
exclude_paths | string | No | Comma-separated regex patterns to skip specific URL paths |
exclude_domains | string | No | Comma-separated regex patterns to block certain domains |
allow_external | boolean | No | Include external domain links in results (default: true) |
include_images | boolean | No | Incorporate images in crawl output |
extract_depth | string | No | Extraction depth: basic (1 credit/5 pages) or advanced (2 credits/5 pages) |
format | string | No | Output format: markdown or text (default: markdown) |
include_favicon | boolean | No | Add favicon URL for each result |
apiKey | string | Yes | Tavily API Key |
| Parameter | Type | Description |
|---|
base_url | string | The base URL that was crawled |
results | array | Array of crawled pages with extracted content |
response_time | number | Time taken for the crawl request in seconds |
request_id | string | Unique identifier for support reference |
Discover and visualize website structure using Tavily's map API. Maps out all accessible URLs from a base URL with depth control, path filtering, and domain restrictions.
| Parameter | Type | Required | Description |
|---|
url | string | Yes | The root URL to begin mapping |
instructions | string | No | Natural language guidance for mapping behavior (costs 2 credits per 10 pages) |
max_depth | number | No | How far from base URL to explore (1-5, default: 1) |
max_breadth | number | No | Links to follow per level (default: 20) |
limit | number | No | Total links to process (default: 50) |
select_paths | string | No | Comma-separated regex patterns for URL path filtering (e.g., /docs/.*) |
select_domains | string | No | Comma-separated regex patterns to restrict mapping to specific domains |
exclude_paths | string | No | Comma-separated regex patterns to exclude specific URL paths |
exclude_domains | string | No | Comma-separated regex patterns to exclude domains |
allow_external | boolean | No | Include external domain links in results (default: true) |
apiKey | string | Yes | Tavily API Key |
| Parameter | Type | Description |
|---|
base_url | string | The base URL that was mapped |
results | array | Array of discovered URLs during mapping |
response_time | number | Time taken for the map request in seconds |
request_id | string | Unique identifier for support reference |
- Category:
tools
- Type:
tavily
