exa

exa

Author: yevanchen Version: 0.0.1 Type: tool

Contact

Description

Exa is an AI-powered search tool that enables semantic search, content retrieval, similarity search, and answers generation using Exa's advanced API.

Tools

1. Exa Search (exa_search)

The search endpoint lets you intelligently search the web and extract contents from the results.

By default, it automatically chooses between traditional keyword search and Exa's embeddings-based model to find the most relevant results for your query.

Parameters:

• query (string, required): The search query to find relevant information on the web.
• search_type (select, optional, default: "neural"):
  • Options: "neural" (semantic), "keyword" (traditional), "auto"
  • Neural uses advanced AI for semantic understanding, keyword uses traditional search techniques
• num_results (number, optional, default: 10): Maximum number of search results (1-100)
• include_domains (string, optional): Comma-separated list of domains to include in results
• exclude_domains (string, optional): Comma-separated list of domains to exclude from results
• start_published_date (string, optional): Only include results published after this date (YYYY-MM-DD)
• end_published_date (string, optional): Only include results published before this date (YYYY-MM-DD)
• use_autoprompt (boolean, optional, default: true): Whether to use Exa's prompt engineering to improve the query
• text_contents (boolean, optional, default: false): Whether to include text contents from each result
• highlight_results (boolean, optional, default: false): Whether to highlight relevant snippets
• category (select, optional): Focus on specific data categories
  • Options: "company", "research paper", "news", "pdf", "github", "tweet", "personal site", "linkedin profile", "financial report"
• includeText (string, optional): Text that must be present in results (up to 5 words)
• excludeText (string, optional): Text that must not be present in results (up to 5 words)

2. Exa Answer (exa_answer)

Get an LLM answer to a question informed by Exa search results. Fully compatible with OpenAI's chat completions endpoint.

/answer performs an Exa search and uses an LLM (GPT-4o-mini) to generate either:

• A direct answer for specific queries (i.e., "What is the capital of France?" would return "Paris")
• A detailed summary with citations for open-ended queries (i.e., "What is the state of AI in healthcare?" would return a summary with citations to relevant sources)

Parameters:

• query (string, required): The question to be answered with supporting evidence from the web
• text (boolean, optional, default: false): Include the full text content of each source in the results
• model (select, optional, default: "exa"):
  • Options: "exa", "exa-pro"
  • Specify which model should process the query and generate the answer

3. Exa Similar Links (exa_similar)

Find similar links to the link provided and optionally return the contents of the pages.

Parameters:

• url (string, required): The source URL to find similar content for
• num_results (number, optional, default: 10): Number of similar links to return (max 100)
• text (boolean, optional, default: false): Include the full text content of each similar page

4. Exa URL Contents (exa_contents)

Get the full page contents, summaries, and metadata for a list of URLs.

Returns instant results from Exa's cache, with automatic live crawling as fallback for uncached pages.

Parameters:

• urls (string, required): Comma-separated list of URLs to extract content from
• livecrawl (select, optional, default: "never"):
  • Options: "never", "fallback", "always", "auto"
  • Choose the live crawling strategy for content retrieval
• full_page_text (boolean, optional, default: false): Include the full text of each webpage, including subpages
• ai_page_summary (boolean, optional, default: false): Generate a summary for each webpage using LLM
• number_of_subpages (number, optional, default: 1): Number of subpages to include in content extraction
• return_links (number, optional, default: 1): Number of links to return from each webpage

Acknowledgements

Special thanks to @ExaAILabs for providing the powerful API that powers this plugin.