Aller au contenu
TakoAPI
Retour aux skills

web-pilot

par liranudi

Web & Frontend Development
4.7kTéléchargements rapportés par clawskills.sh
1Étoiles
0Vues
0 mention J'aimeConnecte-toi pour aimer ce skill
—Pas encore noté
Ta note

Résumé

Search the web and read page contents without API keys.

Description

Ce que fait ce skill

Web browsing toolkit built on four Python scripts. Searches via DuckDuckGo, Brave, or Google across multiple pages, reads page text, runs interactive browser sessions with tab management and click support, and downloads files including PDFs. No API keys required.

Works without any API keys or account signups, using Playwright and Chromium directly to access real browser behavior including cookie banner dismissal.

Quand l'utiliser

  • Scraping search results from DuckDuckGo without an API key
  • Extracting readable text from a paywalled or cookie-gated page
  • Taking a full-page screenshot of a site for visual reference
  • Downloading a PDF and pulling its text content automatically
  • Navigating a multi-tab session to compare content across pages

Exemple de workflow

ENTRÉE

User asks: find recent articles about LLM context windows and summarize the top result

AGENT
  1. 1

    Run google_search.py with the query and --pages 2 to get a result list

  2. 2

    Pick the top URL from the returned JSON

  3. 3

    Run read_page.py on that URL with --format markdown

  4. 4

    Parse the markdown output and extract the main article text

  5. 5

    Return a summary to the user

SORTIE

Markdown text of the article, ready for summarization

Installation

clawhub install liranudi/web-pilot

Exécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.

Liens