Perform real-time web and vertical searches with full-page Markdown extraction.
npx skills add https://github.com/mykcs/myk-skills --skill anysearch-mykcs
Please help me install this Agent Skill. Skill: anysearch Source: https://github.com/mykcs/myk-skills/tree/main/anysearch Command: npx skills add https://github.com/mykcs/myk-skills --skill anysearch-mykcs
AnySearch helps you quickly retrieve up-to-date information from the web and turn relevant pages into usable content, without manually juggling multiple search tools and copy-paste steps.
Ask an AI agent to run an AnySearch extraction for the URL https://example.com/article and return the page content as Markdown.
High-intent search queries and answers about installing and using this skill.
To extract web page content into Markdown for AI research, use the full-page URL extraction feature. It fetches HTML and converts it to Markdown, providing clean text for verification when search snippets are insufficient.
Vertical domain search is a structured search across specific domains like finance, academic, or legal using domain-specific query formats. You need it when researching topics with structured identifiers like DOI, CVE, or ticker symbols to gather authoritative sources efficiently.
Yes, you can run multiple independent search queries in one batch call. This batch searching feature allows you to execute several queries simultaneously, reducing latency and triage time when handling multiple search intents at once.
The search tool requires a JSON-RPC tools/call endpoint and optionally supports Bearer authentication via ANYSEARCH_API_KEY or a .env file. It also requires a local doc-first interface spec step to ensure correct tool parameters and routing for queries.
The best way to discover correct query formats for structured searches is to first use domain listing to identify the compliant format. Once discovered, you can run a vertical search with the correct structured identifiers to gather authoritative sources accurately.
Limitations when using web retrieval for real-time AI agent answers include dependency on the JSON-RPC endpoint availability and the need for a local doc-first interface spec step. Additionally, full-page extraction depends on the target URL's HTML structure being accessible and convertible to Markdown.
anysearch-aiIntegrate a real-time search engine for AI agents via JSON-RPC API.
PancrePal-xiaoyibaoExtract text from local PDF files into structured Markdown format.
Hive-AcademyAutomate Ptah CLI commands for headless agent management and CI pipelines.
DecodoRetrieve structured web data from Google, Amazon, YouTube, and Reddit via the Decodo API.
