webserp
por paperboardofficial
Web & Frontend DevelopmentResumo
Web search across 7 engines in parallel with browser impersonation.
Descrição
O que esta skill faz
webserp queries Google, DuckDuckGo, Brave, Yahoo, Mojeek, Startpage, and Presearch simultaneously from a single command. It uses curl_cffi to mimic browser fingerprints, avoiding blocks that affect plain HTTP clients. Output is SearXNG-compatible JSON with title, URL, content, and source engine per result.
Parallel querying across 7 engines with browser impersonation removes the need for paid search APIs while avoiding the single-engine blind spots and rate limits of direct requests.
Quando usá-la
- Checking release notes for a library version newer than training data
- Finding current news about a security vulnerability or CVE
- Locating official documentation URLs for an unfamiliar framework
- Researching recent events or announcements before writing a summary
- Discovering GitHub repositories or code examples for a specific task
Exemplo de fluxo de trabalho
User asks: What changed in Node.js 22?
- 1
Run
webserp "Node.js 22 release notes" --max-results 5 - 2
Receive JSON output with results array from multiple engines
- 3
Parse results using jq or a JSON library to extract title, url, content fields
- 4
Filter for official nodejs.org or changelog entries
- 5
Return relevant URLs and summaries to the user
A list of recent Node.js 22 changelog entries and official release announcement links with brief content snippets
Instalação
clawhub install paperboardofficial/webserpExecute o comando da CLI ou cole o prompt no Claude Code, Cursor, Codex ou no agente de código que você usa.