Overview
This Skill describes a particular tool route: Tavily and Exa applications exposed through the inference.sh CLI named belt. Installing the Skill gives an agent instructions; it does not itself install every dependency, create a provider account or include service credits. The source examples distinguish finding candidate pages from extracting a known page, so the first decision is whether you have a research question or an existing URL.
For a research task, search for candidate sources, select the relevant results, and extract their actual content before drafting an answer. Preserve the URLs so individual claims remain traceable. The source also illustrates a separate language-model step; its placeholder research results must be replaced with real returned content. Search answers and extraction are intermediate material, not independent confirmation that every claim is true or current.
What to prepare
- A focused query for search, or one or more actual page URLs for extraction.
- A configured belt CLI session and access to the selected inference.sh application.
What you get
- Search results or extracted page content, depending on the selected application.
- A source set that can be passed to a separate summarization or research step.
Suggested workflow
- Review the linked CLI installation instructions and authenticate in your own environment.
- Choose search for discovery or extraction for known URLs, using the application and input shape in the source.
- Inspect the returned pages and dates; replace example placeholders with the actual results before any model handoff.
- Compare important conclusions with the source pages and retain their URLs.
Limits and checks
- The examples use specific application IDs; confirm they are available in the account you will use.
- A subsequent language-model call has its own availability and cost; installing the Skill does not provide it.
- A search snippet is not the complete page, and an extracted page can still contain outdated or unsupported claims.
Common questions
Does installing this Skill give the agent internet access immediately?
It supplies instructions for a CLI workflow. You still need belt, authentication and access to the selected applications in your own environment.
Can I paste the research-to-model example unchanged?
Not as a completed research run. The placeholder must be populated with actual search or extracted content, and the example model application must be available to your account.
Install this skill
Use the Skills CLI to add this skill from its verified GitHub source.
npx skills add https://github.com/101-skills/superpowers --skill web-searchSource information
Rendered from the source repository. Headings, examples, code, tables, links, and referenced images are preserved.
Install the belt CLI skill: npx skills add belt-sh/cliWeb Search & Extraction
Search the web and extract content via inference.sh CLI.
Quick Start
Requires inference.sh CLI (belt). Install instructionsbelt login
# Search the web
belt app run tavily/search-assistant --input '{"query": "latest AI developments 2024"}'Available Apps
Tavily
| App | App ID | Description |
|---|---|---|
| Search Assistant | tavily/search-assistant | AI-powered search with answers |
| Extract | tavily/extract | Extract content from URLs |
Exa
| App | App ID | Description |
|---|---|---|
| Search | exa/search | Smart web search with AI |
| Answer | exa/answer | Direct factual answers |
| Extract | exa/extract | Extract and analyze web content |
Examples
Tavily Search
belt app run tavily/search-assistant --input '{
"query": "What are the best practices for building AI agents?"
}'Returns AI-generated answers with sources and images.
Tavily Extract
belt app run tavily/extract --input '{
"urls": ["https://example.com/article1", "https://example.com/article2"]
}'Extracts clean text and images from multiple URLs.
Exa Search
belt app run exa/search --input '{
"query": "machine learning frameworks comparison"
}'Returns highly relevant links with context.
Exa Answer
belt app run exa/answer --input '{
"question": "What is the population of Tokyo?"
}'Returns direct factual answers.
Exa Extract
belt app run exa/extract --input '{
"url": "https://example.com/research-paper"
}'Extracts and analyzes web page content.
Workflow: Research + LLM
# 1. Search for information
belt app run tavily/search-assistant --input '{
"query": "latest developments in quantum computing"
}' > search_results.json
# 2. Analyze with Claude
belt app run openrouter/claude-sonnet-45 --input '{
"prompt": "Based on this research, summarize the key trends: <search-results>"
}'Workflow: Extract + Summarize
# 1. Extract content from URL
belt app run tavily/extract --input '{
"urls": ["https://example.com/long-article"]
}' > content.json
# 2. Summarize with LLM
belt app run openrouter/claude-haiku-45 --input '{
"prompt": "Summarize this article in 3 bullet points: <content>"
}'Use Cases
- Research: Gather information on any topic
- RAG: Retrieval-augmented generation
- Fact-checking: Verify claims with sources
- Content aggregation: Collect data from multiple sources
- Agents: Build research-capable AI agents
Related Skills
# Full platform skill (all apps)
npx skills add inference-sh/skills@infsh-cli
# LLM models (combine with search for RAG)
npx skills add inference-sh/skills@llm-models
# Image generation
npx skills add inference-sh/skills@ai-image-generationBrowse all apps: belt app list
Documentation
- Adding Tools to Agents - Equip agents with search
- Building a Research Agent - LLM + search integration guide
- Tool Integration Tax - Why pre-built tools matter
