Search, scrape, and monitor web pages through the Firecrawl CLI and REST API.
npx skills add https://github.com/PMN123/trapdoor --skill firecrawl-pmn123
Please help me install this Agent Skill. Skill: firecrawl Source: https://github.com/PMN123/trapdoor/tree/main/.agents/skills/firecrawl Command: npx skills add https://github.com/PMN123/trapdoor --skill firecrawl-pmn123
/search, /scrape, /interact, /parse, /crawl, /monitor), and a smoke-tested setup using FIRECRAWL_API_KEY. - Workflow deliverables: Produce research briefs, SEO audits, lead lists, QA reports, and knowledge bases from collected web evidence. - Use Case: Ask the agent to track a competitor's pricing page; it installs the CLI, creates a recurring monitor with a plain-language goal, and notifies you by email when the page meaningfully changes. ## Quick Start Install Firecrawl and authenticate by asking the agent to run the Firecrawl init command, then scrape a URL you provide into markdown.High-intent search queries and answers about installing and using this skill.
Run `firecrawl scrape <url>` after installing with `npx -y firecrawl-cli@latest init --all --browser`, or call `POST /scrape` on `https://api.firecrawl.dev/v2` with a Bearer API key. Both return clean markdown extracted from the page.
Sign in at firecrawl.dev or run the CLI browser auth flow, which opens a sign-in page and returns the key automatically. Agents can also poll the CLI auth status endpoint with a session ID and PKCE code verifier until the human authorizes.
Yes, the keyless free tier supports search, scrape, interact, parse, and the research index through the official MCP server, CLI, or SDK. It is rate-limited, and crawl, map, monitor, and batch endpoints still require an API key.
Use interact when the page requires clicks, form submissions, or login before content is visible. Scrape handles static extraction of known URLs, while interact drives browser actions on live pages.
Run `firecrawl monitor create` or call `POST /monitor` with a target URL, schedule, and optional plain-language goal. Firecrawl diffs each check against the last snapshot and notifies via webhook, email, or Slack.
The parse endpoint handles local PDF, DOCX, DOC, ODT, RTF, XLSX, XLS, and HTML files up to 50 MB, returning markdown, JSON, or summaries. Documents with public URLs should go through scrape instead.
ironkid90Integrate Firecrawl's search endpoint into workflows for pre-extraction discovery.
ccstudentccSearch biomedical and scientific databases, synthesize evidence, and generate Markdown and PDF reports.
jssfyExtract core motivations, methods, and results from research papers with citations.
finsilabsImplement tiered volume pricing across Shopify, WooCommerce, and BigCommerce.