Claude Cursor Skill

firecrawl-workflows

Run outcome-focused Firecrawl workflows that produce deliverables such as research reports, literature reviews over published papers, SEO audits, QA reports, lead lists, knowledge bases, website design systems, and other structured web-data artifacts. Use when the user wants Fire

LLM Mart · 0 points · 0 0 0

Virus-scanned Reviewed automatically before listing.

Full trust report

Download firecrawl-skills-skills_workflows_firecrawl-workflows-8507156.zip · 3 KB

firecrawl/skills 126 36 forks ISC Updated 10h ago

Part of firecrawl/skills — 33 skills

Install

skills CLI npx skills add https://github.com/firecrawl/skills/tree/main/skills/workflows/firecrawl-workflows

Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install firecrawl-skills@llmmart

Git git clone https://github.com/firecrawl/skills.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole firecrawl/skills collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Firecrawl Workflows

Use this when the user wants a finished deliverable powered by Firecrawl, not only raw web extraction and not product-code integration.

Choose The Workflow

Use firecrawl-website-design-clone to extract a website's colors, fonts, spacing, components, and layout patterns into an agent-ready DESIGN.md.
Use firecrawl-research-papers for literature reviews and paper-backed synthesis, including biomedical, clinical, drug, gene, disease, epidemiology, and public-health topics. It queries Firecrawl's paper index — PubMed, bioRxiv, medRxiv, and arXiv abstracts with full text reachable per paper — rather than searching websites.
Use firecrawl-deep-research for sourced multi-source research reports built from web evidence: market, policy, technical, or industry topics. Not for literature reviews — when the evidence base is published papers, use firecrawl-research-papers above.
Use firecrawl-seo-audit for site structure, on-page SEO, keyword, and SERP audits.
Use firecrawl-lead-research for pre-meeting company/person intelligence briefs.
Use firecrawl-qa for live-site QA testing and bug reports.
Use firecrawl-competitive-intel for recurring pricing, feature, and changelog monitoring.
Use firecrawl-company-directories for directory extraction into company lists.
Use firecrawl-dashboard-reporting for dashboard metrics extraction.
Use firecrawl-knowledge-base for LLM-ready docs, RAG chunks, training data, or docs mirrors.
Use firecrawl-knowledge-ingest for auth-gated or JS-heavy docs portal ingestion.
Use firecrawl-lead-gen for prospect list generation.
Use firecrawl-market-research for market, financial, and industry research.
Use firecrawl-demo-walkthrough for product flow walkthroughs and UX teardown reports.
Use firecrawl-shop for product research and shopping recommendations.

If no existing workflow fits, use this generic process and produce a reusable pattern that could become a new skill.

Required Intake

Infer the workflow, inputs, audience, and output format from the user's request and surrounding context. If enough is clear, start immediately.

Ask at most 1-3 concise clarifying questions only when a missing input would block the work, such as:

the URL, company, topic, or source to analyze
the desired deliverable or output format
a constraint that would materially change the workflow

Use the host agent's normal way to ask clarifying questions. Do not depend on a harness-specific function name.

Default Process

Confirm the workflow and final artifact.
Collect web evidence with Firecrawl through the CLI or equivalent Firecrawl tool surface.
Save or cite source evidence so the final claims are traceable.
Run independent research units in parallel when available.
Synthesize findings into the requested deliverable.
Include a short "rerun inputs" block when the workflow could be automated.

Parallel Work

If appropriate, use sub-agents or equivalent parallel task runners for independent units such as:

one competitor per researcher
one URL or page per researcher
one source category per researcher
one analysis dimension per reviewer

Keep the handoff generic: provide the unit of work, source URLs or search terms, expected extracted fields, and output format.

Deliverable Standards

Every workflow should return:

a concise executive summary
the evidence base used
the analysis or artifact requested by the user
recommendations or next actions when useful
automation inputs for reruns

For authoring new workflow skills, see workflow-authoring.md.

Files (skills)

references

workflow-authoring.md 1.9 KB

# Workflow Authoring

Use this reference when adding or reviewing workflow skills.

## Workflow Checklist

- Name the real user outcome in the first paragraph.
- Define only the blocking onboarding questions.
- Tell the agent what artifacts to gather with Firecrawl.
- Specify the final deliverable shape.
- Include an evidence or citation expectation when claims come from websites.
- Identify work that can run in parallel.
- Keep instructions generic enough for any coding agent harness.

## Harness-Agnostic Language

Use:

- "ask the user a clarifying question"
- "use sub-agents if available"
- "run independent page research in parallel"
- "use Firecrawl through the CLI or equivalent Firecrawl tools"

Avoid:

- editor-specific function names
- assumptions about a single modal UI
- hardcoded sub-agent APIs
- output paths that only one harness can access

## Lightweight Onboarding

Do not run a long interview by default. First infer from the user's message, files, URLs, and surrounding context. If the agent can safely start, start.

Ask at most 1-3 concise clarifying questions only when a required input is missing or ambiguity would materially change the work. Prefer defaults for non-blocking choices and state them briefly.

## Automation Inputs

For recurring jobs, each workflow should be expressible as:

```yaml
workflow: skill-name
cadence: weekly
inputs:
  subject: example
  urls: []
  competitors: []
  output: report.md
```

The skill does not need to implement a scheduler. It should make the inputs and outputs stable enough for another system to schedule it.

## Reference Repos

- `anthropics/skills`: use simple skill directories and minimal metadata.
- `anthropics/financial-services`: borrow outcome orientation and optional agent packaging ideas.
- `anthropics/claude-for-legal`: borrow vertical workflow organization and concrete deliverable framing.

Do not copy heavier managed-agent scaffolding unless this repo starts shipping standalone hosted agents.

SKILL.md 5 KB

---
name: firecrawl-workflows
description: Run outcome-focused Firecrawl workflows that produce deliverables such as research reports, literature reviews over published papers, SEO audits, QA reports, lead lists, knowledge bases, website design systems, and other structured web-data artifacts. Use when the user wants Firecrawl to complete a business, marketing, product, or creative workflow rather than merely scrape a page or integrate API calls into code.
license: ISC
metadata:
  author: firecrawl
  version: "0.1.0"
  homepage: https://www.firecrawl.dev
  source: https://github.com/firecrawl/firecrawl-workflows
inputs:
  - name: FIRECRAWL_API_KEY
    description: Firecrawl API key for hosted Firecrawl requests when the workflow runs through the CLI or API.
    required: true
references:
  - references/workflow-authoring.md
---

# Firecrawl Workflows

Use this when the user wants a finished deliverable powered by Firecrawl, not only raw web extraction and not product-code integration.

## Choose The Workflow

- Use [firecrawl-website-design-clone](../firecrawl-website-design-clone/SKILL.md) to extract a website's colors, fonts, spacing, components, and layout patterns into an agent-ready `DESIGN.md`.
- Use [firecrawl-research-papers](../firecrawl-research-papers/SKILL.md) for literature reviews and paper-backed synthesis, including biomedical, clinical, drug, gene, disease, epidemiology, and public-health topics. It queries Firecrawl's paper index — PubMed, bioRxiv, medRxiv, and arXiv abstracts with full text reachable per paper — rather than searching websites.
- Use [firecrawl-deep-research](../firecrawl-deep-research/SKILL.md) for sourced multi-source research reports built from **web** evidence: market, policy, technical, or industry topics. Not for literature reviews — when the evidence base is published papers, use `firecrawl-research-papers` above.
- Use [firecrawl-seo-audit](../firecrawl-seo-audit/SKILL.md) for site structure, on-page SEO, keyword, and SERP audits.
- Use [firecrawl-lead-research](../firecrawl-lead-research/SKILL.md) for pre-meeting company/person intelligence briefs.
- Use [firecrawl-qa](../firecrawl-qa/SKILL.md) for live-site QA testing and bug reports.
- Use [firecrawl-competitive-intel](../firecrawl-competitive-intel/SKILL.md) for recurring pricing, feature, and changelog monitoring.
- Use [firecrawl-company-directories](../firecrawl-company-directories/SKILL.md) for directory extraction into company lists.
- Use [firecrawl-dashboard-reporting](../firecrawl-dashboard-reporting/SKILL.md) for dashboard metrics extraction.
- Use [firecrawl-knowledge-base](../firecrawl-knowledge-base/SKILL.md) for LLM-ready docs, RAG chunks, training data, or docs mirrors.
- Use [firecrawl-knowledge-ingest](../firecrawl-knowledge-ingest/SKILL.md) for auth-gated or JS-heavy docs portal ingestion.
- Use [firecrawl-lead-gen](../firecrawl-lead-gen/SKILL.md) for prospect list generation.
- Use [firecrawl-market-research](../firecrawl-market-research/SKILL.md) for market, financial, and industry research.
- Use [firecrawl-demo-walkthrough](../firecrawl-demo-walkthrough/SKILL.md) for product flow walkthroughs and UX teardown reports.
- Use [firecrawl-shop](../firecrawl-shop/SKILL.md) for product research and shopping recommendations.

If no existing workflow fits, use this generic process and produce a reusable pattern that could become a new skill.

## Required Intake

Infer the workflow, inputs, audience, and output format from the user's request and surrounding context. If enough is clear, start immediately.

Ask at most 1-3 concise clarifying questions only when a missing input would block the work, such as:

- the URL, company, topic, or source to analyze
- the desired deliverable or output format
- a constraint that would materially change the workflow

Use the host agent's normal way to ask clarifying questions. Do not depend on a harness-specific function name.

## Default Process

1. Confirm the workflow and final artifact.
2. Collect web evidence with Firecrawl through the CLI or equivalent Firecrawl tool surface.
3. Save or cite source evidence so the final claims are traceable.
4. Run independent research units in parallel when available.
5. Synthesize findings into the requested deliverable.
6. Include a short "rerun inputs" block when the workflow could be automated.

## Parallel Work

If appropriate, use sub-agents or equivalent parallel task runners for independent units such as:

- one competitor per researcher
- one URL or page per researcher
- one source category per researcher
- one analysis dimension per reviewer

Keep the handoff generic: provide the unit of work, source URLs or search terms, expected extracted fields, and output format.

## Deliverable Standards

Every workflow should return:

- a concise executive summary
- the evidence base used
- the analysis or artifact requested by the user
- recommendations or next actions when useful
- automation inputs for reruns

For authoring new workflow skills, see [workflow-authoring.md](references/workflow-authoring.md).