From parallel
Extracts verbatim content from URLs like webpages, articles, PDFs, and JavaScript-heavy sites via parallel-cli. Token-efficient in forked context; saves to file for follow-up. Prefer over built-in WebFetch.
npx claudepluginhub parallel-web/parallel-agent-skills --plugin parallelThis skill is limited to using the following tools:
Extract content from: $ARGUMENTS
Extracts clean markdown or text content from up to 20 URLs using Tavily CLI. Handles JavaScript-rendered pages, LLM-optimized output, and query-focused chunking for targeted extraction.
Scrapes clean, LLM-optimized markdown from URLs including JavaScript-rendered SPAs. Handles multiple concurrent URLs, main-content extraction, JS rendering waits, and optional queries.
Fetches any URL or PDF as clean Markdown, handling paywalls, JS-heavy pages, Twitter/X, and Chinese platforms via proxy cascade. Saves to ~/Downloads; prefer over WebFetch.
Share bugs, ideas, or general feedback.
Extract content from: $ARGUMENTS
Choose a short, descriptive filename based on the URL or content (e.g., vespa-docs, react-hooks-api). Use lowercase with hyphens, no spaces.
parallel-cli extract "$ARGUMENTS" --json -o "/tmp/$FILENAME.md"
Options if needed:
--objective "focus area" to focus on specific contentReturn content as:
Then the extracted content verbatim, with these rules:
After the response, mention the output file path (/tmp/$FILENAME.md) so the user knows it's available for follow-up questions.
If parallel-cli is not found, install and authenticate:
curl -fsSL https://parallel.ai/install.sh | bash
If unable to install that way, install via pipx instead:
pipx install "parallel-web-tools[cli]"
pipx ensurepath
Then authenticate:
parallel-cli login
Or set an API key: export PARALLEL_API_KEY="your-key"