web-research_skill

This skill enables deep web research by using Exa and Firecrawl endpoints to find sources, extract clean text, and answer factual questions.

2

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill merit-systems/x402scan-skills --skill web-research

SKILL.md7.1 KB

Overview

This skill provides neural web search and robust content extraction through x402-protected APIs. It combines semantic search (Exa) with a resilient scraper (Firecrawl) to handle blocked or JS-heavy sites and to return clean, LLM-ready text and direct factual answers. Use it when standard fetchers fail or when you need conceptual matches and multi-source synthesis.

How this skill works

You interact with x402 endpoints that proxy Exa (neural search, find-similar, contents, answer) and Firecrawl (scrape, search). Before using any endpoint you must discover or copy the exact API paths under the enrichx402.com /api/ prefix. Exa performs semantic ranking and content extraction; Firecrawl renders and scrapes pages that block conventional fetchers or require JS.

When to use it

Deep investigations that need multiple corroborating sources
Finding pages similar to a reference URL (competitor or related content discovery)
Extracting clean, structured text from a list of URLs
Scraping paywalled, bot-blocked, or JavaScript-heavy pages
Getting quick factual answers with source citations

Best practices

Always call x402.discover_api_endpoints(url="https://enrichx402.com") or copy exact /api/ URLs before fetching
Prefer Exa for semantic queries and bulk content extraction (exa/contents) to save cost
Try free WebSearch/WebFetch first; fallback to Exa/Firecrawl only when needed
Batch URL extractions in a single exa/contents call for cost efficiency
Use Firecrawl scrape with waitFor for pages that require JS rendering or are blocked

Example use cases

Researching market landscape: run Exa search, find-similar on top hits, extract with exa/contents, synthesize
Investigative reporting: scrape a blocked page with firecrawl/scrape and extract main content as markdown
Source expansion: provide a reference URL to exa/find-similar to discover related articles and competitors
Fact-checking: query exa/answer for direct answers with source citations
Bulk content ingestion: pass many URLs to exa/contents to obtain clean text for downstream LLM analysis

FAQ

No. Never guess paths. Use the exact enrichx402.com /api/ URLs or call x402.discover_api_endpoints to obtain them.

When should I use Firecrawl versus Exa?

Use Exa for semantic search, similarity, and cheap text extraction. Use Firecrawl when pages are blocked, require JS rendering, or you need scraper-optimized markdown.