Firecrawl Research Index Skill for Claude Code: What It Does and How to Install It

Makes your assistant find research papers in Firecrawl's paper index, then expand to related papers and check claims in the paper text.

Installs: 24.8K; on skills.sh
Stars: 142; on GitHub
Updated: Oct 2026; last repo commit

SKILL.md tells your assistant to read the question, pick an approach and run firecrawl research commands (or the matching firecrawl_research_* MCP tools). It lists seven query shapes, from one named paper to leaderboard questions, and says to lean toward returning the fuller set of relevant papers.

It loads when a task is about finding papers. It needs the Firecrawl CLI or the Firecrawl MCP server. The skill does not bundle scripts.

Good for

Researchers doing literature searches in biomedical, life-science, CS, physics or math topics
Developers building a reading list on a method, benchmark or dataset
Anyone who wants papers and their citers or references, not web pages

Install firecrawl-research-index

Claude Code

Copy this prompt.

Install the firecrawl-research-index skill in Claude Code. Run this command:

npm install -g firecrawl-cli@latest
firecrawl login --browser
npx skills add https://github.com/firecrawl/skills --skill firecrawl-research-index

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.

Requires Node.js, A Firecrawl account.

The skill only gives your assistant instructions. The Firecrawl command line tool (or the Firecrawl MCP server) does the work, so install the CLI first and sign in with firecrawl login --browser, or set FIRECRAWL_API_KEY. The plugin route installs the repo's firecrawl plugin: the Claude plugin.json lists 17 skills (12 core and 5 build), and the repo root has a .mcp.json that starts the firecrawl-mcp server with npx. The README says the plugin includes Firecrawl MCP configuration. There is no Claude app upload route because the skill drives a local CLI or MCP tools.

From the vendor's docs. Source.

What's in the skill

When it loads

Find the papers that answer a research query in Firecrawl's research paper index — a corpus of paper abstracts whose largest share is biomedical and life-science literature (PubMed, bioRxiv, medRxiv), alongside arXiv preprints in CS, physics, and math — using semantic search, semantic and structural expansion, and in-body verification. Use this skill for literature-finding and paper-retrieval tasks of any kind, including clinical, biomedical, drug, gene, disease, and other life-science questions, whether the answer is a single paper or a full multi-paper set. The index is reached only through the `firecrawl_research_*` MCP tools or the `firecrawl research` CLI subcommands. Calling `firecrawl_search` with its `categories` option set to `["research"]` is a different feature — it filters ordinary web search to research-affiliated websites (the list includes PubMed, bioRxiv, medRxiv, arXiv, and publisher sites) and returns page results from them, without querying the paper records in this index.

Sections in SKILL.md

What is in the index
The tools, and what each is uniquely good at
Match the approach to the query
Principles
See also

Key facts

Author: Firecrawl (firecrawl/skills catalog)
Repo: firecrawl/skills
License: ISC (LICENSE and README agree)
Installs on skills.sh: 24,821 (Nov 2, 2026)
GitHub stars: 142 for the whole firecrawl/skills repo (Nov 2, 2026)
Last repo commit: Oct 22, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: warn
What it includes: 1 file: SKILL.md (counted from the repo tree)
Needs: The Firecrawl CLI or the Firecrawl MCP server, plus Node.js to install the CLI
Part of plugin: firecrawl (listed in the Claude plugin.json)
Source of truth: The README says the skills/ folders in this repo are CI-synced mirrors and that core skills are maintained in firecrawl/cli

What it includes

Semantic search over paper abstracts, the first move for most queries.
Finds papers that match a question in plain words. If results look thin, rerun with a different framing.
Try: “Find papers on training-free detection of AI-written text.”
Network · firecrawl research search-papers
Expands from seed papers to similar papers, citers or references, ranked to your intent.
Turns one good paper into the rest of a set. Modes are similar, citers and references.
Network · firecrawl research related-papers
Returns title, abstract, authors, categories, source ids and dates for one paper.
Use it to get a full citation for a candidate. It does not read the paper body.
Network · firecrawl research inspect-paper
Returns passages from the body of one paper that answer a question.
Settles a specific doubt, such as a score a paper reports or a method it uses.
Try: “Check what read-paper says about the dataset used.”
Network · firecrawl research read-paper
A web search limited to research sites. SKILL.md says it is not this index.
The skill points here only when you are already running a web search. It returns ordinary web results.
Network · firecrawl search --categories research
General web search and page fetch for leaderboards and rankings.
Used for superlative questions such as best on a benchmark, then each top entry is mapped back to a paper.
Network · firecrawl search / firecrawl scrape

From the official docs

Permissions

Low risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Your query text goes to Firecrawl
Paper searches are sent to the Firecrawl API. Leave confidential details out of queries. github.com
Setup installs a global npm package
The CLI is installed with npm install -g, and install.md says the one-command setup adds skills to every detected editor. The single-skill route in the install box avoids that. github.com
The plugin route adds an MCP config
The repo root has a .mcp.json that starts firecrawl-mcp with npx -y and reads FIRECRAWL_API_KEY. The README says the plugin includes Firecrawl MCP configuration. Use a skills route if you only want this skill. github.com

Technical details

Turning on read-only	No read-only switch is documented.
Talks to	Firecrawl's API (api.firecrawl.dev)
Sign-in and keys	firecrawl login --browser stores credentials in the CLI (install.md). You can set FIRECRAWL_API_KEY instead.
Scopes	Runs firecrawl commands in your shell, or calls Firecrawl MCP tools Sends your query text to Firecrawl

Questions

What does the firecrawl-research-index skill do?

It makes your assistant search Firecrawl's research paper index, which SKILL.md describes as paper abstracts, mostly PubMed, bioRxiv and medRxiv, plus arXiv preprints. It then expands from good hits to related papers and checks claims inside a paper body.

How do I install it?

Use the install box at the top of this page. You need the Firecrawl CLI installed and signed in, because the skill reaches the index through firecrawl research commands or Firecrawl MCP tools.

Do I need a Firecrawl account or API key?

Install the CLI and sign in with firecrawl login --browser, or set FIRECRAWL_API_KEY. SKILL.md does not say whether the research commands also run without an account, so check the CLI docs before relying on that.

Is it safe?

It runs Firecrawl commands in your shell and sends your search text to Firecrawl, so avoid putting confidential details in queries. It bundles no scripts, only SKILL.md. Our advice: review the commands your assistant proposes before approving them.

How is it different from a normal web search for papers?

SKILL.md says the index supports semantic search over abstracts, related-paper expansion, canonical metadata and in-body passages, which a web search filtered to research sites does not. It calls firecrawl search --categories research a different feature that returns ordinary web results.

Does it cover paywalled or niche papers?

Not always. SKILL.md says coverage outside PubMed, bioRxiv, medRxiv and arXiv is thinner, and a paper that exists only behind a publisher paywall or in a niche venue may not be indexed.

Is it official?

Yes. It is in the vendor's own repo, firecrawl/skills (owner Firecrawl). The README calls the repo a distribution catalog and says core skills, including this one, are maintained in firecrawl/cli and mirrored here.

Does it work in the Claude app?

The install box has no Claude app upload route, because the skill depends on the Firecrawl CLI or MCP tools running on your machine.

Information

updated Nov 2, 2026

Repository: firecrawl/skills
Folder: skills/core/firecrawl-research-index

firecrawl/skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 6

firecrawl/skills README · github.com
firecrawl/skills marketplace.json · github.com
firecrawl/skills plugin.json · github.com
Firecrawl CLI install.md · github.com
firecrawl/skills .mcp.json · github.com
firecrawl-research-index SKILL.md · github.com