firecrawl-deep-research Skill for Claude: What It Does and How to Install It
Makes Claude run a deep web research job with Firecrawl and write a cited report with findings, risks and open questions.
- Installs
- 35.7K
- on skills.sh
- Stars
- 214
- on GitHub
- Updated
- Nov 2026
- last repo commit
This skill produces a formal, cited research report from the open web. It first asks how long you want the research to run, then searches and scrapes with Firecrawl at that depth: a few queries and sources for a quick run, up to 10 queries and 25 sources for a thorough one, and 10+ queries and 25+ sources for an exhaustive one.
It needs a Firecrawl API key, which the skill's header marks as required, and Firecrawl search and scrape through the CLI or an equivalent tool. It is for report-scale work only: SKILL.md says not to use it for product picks, top-N lists or quick lookups, and to use firecrawl-research-papers for literature reviews.
Good for
- Analysts who need a formal report on a complex topic
- Technical, policy and market researchers
- Writers who need every claim traced to a source
Install firecrawl-deep-research
Claude Code
- 1
Copy this prompt.
Install the firecrawl-deep-research skill in Claude Code. Run this command: /plugin marketplace add firecrawl/firecrawl-workflows /plugin install firecrawl-workflows@firecrawl-workflows Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Pick one route. The README prefers the full firecrawl/skills catalog, and also says installing from firecrawl/firecrawl-workflows works and serves the same 19 workflow skills. The plugin route installs the firecrawl-workflows plugin, whose plugin.json lists 16 skills and which has no hook or MCP config files in the repo tree. The skill needs a Firecrawl API key and Firecrawl search and scrape tools, so the box has no Claude app upload.
Built from the package data we hold. Not yet tested by us. Source.
What's in the skill
When it loads
Produce an intensive, cited analytical report: executive summary, multi-angle findings, contrarian views, open questions, and full sources. Use only when the user needs rigorous synthesis of a complex topic (scientific, technical, policy, or market-analytical) that cannot be answered with a short search, and wants a formal written report, not a recommendation list. Do not use for product picks, top-N lists, quick lookups, or routine "find out about X" tasks. If the request does not clearly need this kind of report, do not use this skill. Do not use for a literature review over published papers. This skill collects evidence from the open web. A request for the literature on a biomedical, clinical, life-science, or other scientific topic — papers, studies, trials, preprints — belongs to firecrawl-research-papers, which queries Firecrawl's paper index (PubMed, bioRxiv, medRxiv, arXiv) instead of searching websites.
Sections in SKILL.md
- Onboarding Interview
- Firecrawl Collection Plan
- When Published Papers Are The Evidence
- Parallel Work
- Final Deliverable
- Executive Summary
- Key Findings
- Detailed Analysis
- Contrarian Views And Risks
- Open Questions
- Sources
- Rerun Inputs
- Quality Bar
Key facts
- Installs on skills.sh
- 35,661 (Nov 2, 2026)
- Last repo commit
- Nov 1, 2026
- What it includes
- 1 file: SKILL.md only (no scripts)
- Skill version
- 0.1.0 (SKILL.md metadata)
What it includes
Asks how long the research should run
Maps a few minutes to Quick, 10 to 15 minutes to Thorough, and longer to Exhaustive.
Try: “Do a deep research report on the state of solid-state batteries. Take 15 minutes.”
Read ·
Onboarding question3 to 5 queries and 5 to 10 scraped sources
Collects with Firecrawl search and scrape through the CLI or equivalent tool.
Network ·
Quick depth5 to 10 queries and 15 to 25 scraped sources
Searches from different angles, and avoids re-scraping URLs already returned with full content.
Network ·
Thorough depth10+ queries and 25+ sources
Includes primary sources, research papers, expert views and contrarian sources.
Network ·
Exhaustive depthSub-agents by research angle, if appropriate
Angles are overview, technical detail, market context, contrarian views and primary sources. Each returns claims, URLs, quality notes and uncertainty.
Read ·
Parallel workLiterature reviews go to firecrawl-research-papers
If the evidence is published papers, SKILL.md sends the request to that skill instead.
Read ·
Hand-off for papersExecutive summary, findings, analysis, risks, questions, sources
Ends with a Rerun Inputs block listing workflow, topic, depth and output format.
Write ·
Final reportCite claims, prefer primary sources, flag conflicts
Synthesize instead of listing scrape summaries.
Read ·
Quality bar
From the official docs
Permissions
Medium risk
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It needs your Firecrawl API key
The skill header marks FIRECRAWL_API_KEY as required. SKILL.md does not say what a run costs, so check your Firecrawl plan before picking Exhaustive. github.com
Depth sets how much it fetches
Quick uses 3 to 5 queries, Thorough 5 to 10, and Exhaustive 10 or more, with 25 or more scraped sources. Start with Quick if you are unsure. github.com
The plugin route installs more than this skill
The firecrawl-workflows plugin.json lists 16 skills and the repo tree shows no hook or MCP config files for it. The skills CLI single-skill command installs only this one. github.com
Security advisories
No published security advisories as of Nov 2, 2026.
Technical details
| Turning on read-only | No read-only switch. The skill bundles no scripts. |
|---|---|
| Sign-in and keys | FIRECRAWL_API_KEY is required per the SKILL.md header. It does not say where the key is stored. |
| Scopes |
|
Questions
What does the firecrawl-deep-research skill do?
It runs a multi-source research job on the open web with Firecrawl and writes a formal cited report. The report has an executive summary, findings, contrarian views, open questions and a full source list.
How do I install firecrawl-deep-research?
Use the install box at the top of this page. The README prefers the full firecrawl/skills catalog. The box also has a single-skill command and a plugin route, and it needs a Firecrawl API key.
Do I need a Firecrawl account?
Yes. The skill header marks FIRECRAWL_API_KEY as required, and SKILL.md tells Claude to use Firecrawl search and scrape through the CLI or an equivalent tool. SKILL.md does not give the cost of a run.
How long does a run take?
You choose. Claude asks first. A few minutes means Quick, about 10 to 15 minutes means Thorough, and anything longer means Exhaustive.
When should I use firecrawl-research-papers instead?
For literature reviews and biomedical, clinical or other scientific topics where the answer lives in papers. SKILL.md says that skill queries Firecrawl's paper index, while this one searches websites.
Is it only for long reports?
Yes. SKILL.md says not to use it for product picks, top-N lists, quick lookups or routine "find out about X" tasks.
Is it safe?
It bundles no scripts, but it uses your Firecrawl API key to search and scrape many web pages, and may start sub-agents. Keep the key in an environment variable, and pick a lower depth first (our advice).
Information
updated Nov 2, 2026
- Repository
- firecrawl/firecrawl-workflows
- Folder
skills/firecrawl-deep-research
Sources
Docs, issue trackers and discussions this page is built from.
Show all 4Hide
- firecrawl-deep-research SKILL.md · github.com
- firecrawl-workflows README · github.com
- firecrawl-workflows marketplace.json · github.com
- firecrawl-workflows plugin.json · github.com