firecrawl-deep-research

Generates cited deep-research reports from open-web sources via Firecrawl search and scrape.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/leonardoacosta/agents --skill firecrawl-deep-research-leonardoacosta
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: firecrawl-deep-research
Source: https://github.com/leonardoacosta/agents/tree/main/skills/firecrawl-deep-research
Command: npx skills add https://github.com/leonardoacosta/agents --skill firecrawl-deep-research-leonardoacosta

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Complex scientific, technical, policy, or market questions often require synthesizing dozens of web sources into a rigorous, cited report, which is slow and error-prone to do manually. This Skill automates that end-to-end research workflow using Firecrawl's search and scrape capabilities. ## Core Features & Use Cases - Depth-Tiered Collection: Maps the user's available runtime to Quick, Thorough, or Exhaustive plans that scale the number of search queries and scraped sources. - Structured Report Output: Produces a formal Markdown report with executive summary, key findings, detailed analysis, contrarian views, open questions, and a full source list. - Literature Handoff: Detects when the evidence base is published papers and delegates to firecrawl-research-papers, which queries Firecrawl's paper index (PubMed, bioRxiv, medRxiv, arXiv). - Use Case: Ask for a thorough analysis of a new regulatory policy's market impact; the Skill searches from multiple angles, scrapes 15-25 sources, and delivers a cited report with risks and open questions. ## Quick Start Ask the agent to produce a deep research report on your chosen topic and specify how long the research should run.

Frequently Asked Questions about firecrawl-deep-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a cited deep research report with Firecrawl?

Invoke the skill with your topic and state how long the research should run. It maps your answer to a Quick, Thorough, or Exhaustive plan, runs Firecrawl search and scrape across multiple angles, and returns a structured Markdown report with cited sources.

What is the difference between firecrawl-deep-research and firecrawl-research-papers?

firecrawl-deep-research collects evidence from the open web using search and scrape. firecrawl-research-papers queries Firecrawl's paper index (PubMed, bioRxiv, medRxiv, arXiv) and is the right choice for literature reviews or biomedical and scientific topics.

Does Firecrawl search with the research category query academic papers?

No. Passing categories: ["research"] filters an ordinary web search to research-affiliated websites and returns their web pages, not paper records. Use firecrawl-research-papers to query the actual paper index with abstracts and full text.

When should I not use deep web research for a question?

Avoid it for product picks, top-N lists, quick lookups, or routine questions answerable with a short search. The skill is designed only for report-scale synthesis of complex topics that the user explicitly wants delivered as a formal written report.

What API key do I need for Firecrawl deep research?

You need a Firecrawl API key set as the FIRECRAWL_API_KEY environment variable. It authenticates hosted Firecrawl search and scrape requests made through the CLI or MCP tool surface during collection.