exa-search
Exa Search performs semantic web search and content discovery through embeddings, supporting multiple query types including similarity matching, category-specific searches, and structured research output. Route queries to the appropriate endpoint—search for general topics, findsimilar for related pages, contents for full text extraction, answer for direct responses, or research for formatted findings.
Exa Search finds semantically similar web pages and content using embeddings-based search across multiple categories.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-04-21
Exa Search finds semantically similar web pages and content using embeddings-based search across multiple categories. Exa Search performs semantic web search and content discovery through embeddings, supporting multiple query types including similarity matching, category-specific searches, and structured research output. Route queries to the appropriate endpoint—search for general topics, findsimilar for related pages, contents for full text extraction, answer for direct responses, or research for formatted findings.
Use it when
- Yes, exa-search is built on embeddings-based semantic search technology.
- Exa Search excels at discovering research papers, code repositories, domain-specific content, and general web pages.
Verify before relying
Read SKILL.md below before installing (5 files). Open directory: indexed for reading, not audited.
Install
BenedictKing/exa-search/exa-search · repository language: JavaScript
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How does exa-search find similar web pages?
Exa Search uses embeddings-based semantic matching to find similar web pages. When you query for related content, exa-search analyzes the semantic meaning rather than just keywords, allowing it to discover pages with conceptually similar content even if they use different terminology. This embeddings approach powers the findsimilar endpoint for content discovery.
Can exa-search perform semantic search using embeddings?
Yes, exa-search is built on embeddings-based semantic search technology. The skill routes queries intelligently—use the search endpoint for general topics, findsimilar for embeddings-based similarity matching, research for structured categorized results, contents for full text extraction, or answer for direct responses with citations. This multi-endpoint design lets you choose the semantic approach that fits your need.
What types of content can exa-search find?
Exa Search excels at discovering research papers, code repositories, domain-specific content, and general web pages. The research endpoint returns structured, categorized findings ideal for academic work. The skill also supports category-specific searches, allowing you to filter results by content type. Use the contents endpoint to extract and summarize full text from discovered sources.
How do I use exa-search for structured research queries?
Exa Search's research endpoint delivers structured research output with categorized results, making it ideal for organized investigations. This endpoint formats findings in a way that supports systematic exploration of topics. Combine it with the answer endpoint to get direct responses with cited sources, or use findsimilar to discover related research papers and articles on your topic.
What is the exa-search skill's license?
Exa Search is released under the MIT license, making it freely available for both commercial and personal use with minimal restrictions.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
Exa Search Skill
Trigger Conditions & Endpoint Selection
Choose Exa endpoint based on user intent:
- search: Need semantic search / find web pages / research topics
- contents: Given result IDs, need to extract full content
- findsimilar: Given URL, need to find similar pages
- answer: Need direct answer to a question
- research: Need structured research output following given
output_schema
Recommended Architecture (Main Skill + Sub-skill)
This skill uses a two-phase architecture:
- Main skill (current context): Understand user question → Choose endpoint → Assemble JSON payload
- Sub-skill (fork context): Only responsible for HTTP call execution, avoiding conversation history token waste
Execution
(truncated - see the full file via the links below)
File tree — 5 files
.claude/skills/exa-search/.env.example
.claude/skills/exa-search/.gitignore
.claude/skills/exa-search/SKILL.md
.claude/skills/exa-search/exa-api.cjs
.claude/skills/exa-search/exa-fetcher.md
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Find semantically similar web pages and content using embeddings”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
exa-search connects to Exa's MCP endpoint for live web queries, code documentation lookup, and full-page retrieval. Use the simple search tool for basic queries or the advanced variant for filtered results by date, domain, category, or content. Fetch complete pages when search highlights aren't enough.
Exa Search integrates Exa's Search API to conduct web research with fine-grained control over results. Choose your search mode, apply domain and date filters, and request only the content you need—highlights for efficiency or full text for deep reading. Built-in workflows handle news lookups, source verification, and citation management.
web-search-exa provides neural search powered by embeddings to discover relevant web content, extract full page text, and synthesize findings. Query the web intelligently across multiple search modes, filter by domain or date, and get LLM-generated answers informed by live search results.
This skill leverages the Firecrawl API to pull content from web pages in multiple formats—markdown, HTML, JSON, screenshots, and PDFs. It supports single-page scraping, full-site crawling with depth and path controls, batch processing of multiple URLs, and browser automation through actions like clicks and scrolls. Use it whenever you need to gather web data, extract structured information with custom schemas, or interact with dynamic pages.
Tavily Web integrates the Tavily API to handle multiple research workflows: search the web for current information, extract and summarize content from specific URLs, crawl sites to discover page structure, and generate structured research output. Route requests to the appropriate endpoint based on user intent—search for latest sources, extract for URL content, crawl for site traversal, map for page discovery, or research for schema-driven results.
This skill provides patterns for integrating Exa.ai's search API into applications using Python (exa_py) or TypeScript (exa-js). It covers search modes—auto, neural, and keyword—along with domain and date filtering, text extraction, summaries, and highlights to help you retrieve and process web content effectively.
More skills Web Research (unlicensed) · Exa Search (unlicensed) · exa-rag (MIT) · exa-search (MIT) · exa-entities (MIT)