Firecrawl Workflows — Setup Guide¶

Source: firecrawl/firecrawl-workflows (120K+ combined installs) Category: Growth & Research Languages: TypeScript

Production-grade web research and growth workflows from Firecrawl. Four high-value workflows purpose-built for AI agents: deep research (multi-page synthesis with citations), lead generation (automated discovery and enrichment), market research (competitive landscape analysis), and SEO audit (technical and content analysis). All built on Firecrawl's infrastructure with rate limiting, proxy rotation, and JavaScript rendering.

Installation¶

# Deep Research — multi-page research synthesis
npx skills add firecrawl/firecrawl-workflows@firecrawl-deep-research

# Lead Generation — automated lead discovery and enrichment
npx skills add firecrawl/firecrawl-workflows@firecrawl-lead-gen

# Market Research — competitive landscape analysis
npx skills add firecrawl/firecrawl-workflows@firecrawl-market-research

# SEO Audit — technical and content SEO analysis
npx skills add firecrawl/firecrawl-workflows@firecrawl-seo-audit

# Knowledge Base — build searchable knowledge bases from websites (bonus)
npx skills add firecrawl/firecrawl-workflows@firecrawl-knowledge-base

Verify:

npx skills list | grep firecrawl

Prerequisites¶

Requirement	Details
Firecrawl API Key	Sign up at firecrawl.dev — free tier: 500 credits/month
Hermes Agent	Any version
Node.js	18+

Set API key:

export FIRECRAWL_API_KEY="fc-..."

Workflow Details¶

1. firecrawl-deep-research (30.5K installs)¶

Multi-page research with automatic sub-question generation, source tracking, and structured output.

firecrawl-deep-research "What are the top MCP server frameworks in 2026?" \
    --depth 3 \
    --sources 10 \
    --format markdown

Parameter	Default	Description
`--depth`	2	How many levels of follow-up questions
`--sources`	5	Minimum unique sources to gather
`--format`	markdown	Output format: markdown, json, or html
`--citations`	true	Include inline source citations

Output structure:

# Research: MCP Server Frameworks 2026

## Executive Summary
...

## Key Findings
1. **Finding one** (see sources below)
2. **Finding two** (see sources below)

## Detailed Analysis
...

## Sources
1. Firecrawl web extraction documentation — Relevance: High
...

2. firecrawl-lead-gen (29.1K installs)¶

Automated lead discovery with domain filtering and contact enrichment.

firecrawl-lead-gen \
    --industry "e-commerce" \
    --company-size "10-200" \
    --role "CTO, VP Engineering" \
    --domains "shopify.com,bigcommerce.com" \
    --max-leads 50

Parameter	Default	Description
`--industry`	(required)	Target industry
`--company-size`	any	Employee count range
`--role`	any	Target job titles
`--domains`	(optional)	Limit to specific domains
`--enrich`	true	Find email, LinkedIn, company data
`--max-leads`	25	Cap results

Output: CSV with columns: Company, Contact Name, Title, Email, LinkedIn, Company Size, Funding, Tech Stack.

3. firecrawl-market-research (29.4K installs)¶

Competitive landscape mapping with feature comparison and pricing analysis.

firecrawl-market-research \
    --company "CorpusIQ" \
    --competitors "Glean, Notion AI, Mem, Tana" \
    --dimensions "features,pricing,positioning,gtm"

Parameter	Default	Description
`--company`	(required)	Your company name
`--competitors`	(required)	Comma-separated competitor list
`--dimensions`	all	What to analyze
`--depth`	2	Research depth per competitor

Output structure: - Feature comparison matrix - Pricing tier comparison - Positioning/messaging analysis - GTM strategy comparison - Strengths/weaknesses per competitor

4. firecrawl-seo-audit (29.2K installs)¶

Technical SEO crawl with content gap analysis.

firecrawl-seo-audit https://corpusiq.io \
    --competitors "glean.com,notion.so" \
    --check "technical,content,backlinks"

Check Type	What It Covers
technical	Broken links, redirects, metadata, schema, Core Web Vitals
content	Keyword gaps, thin content, duplicate content, readability
backlinks	Domain authority, referring domains, anchor text distribution

Hermes/CorpusIQ Relevance¶

Growth Operations¶

firecrawl-lead-gen automates prospect discovery for CorpusIQ outreach campaigns. Replace manual LinkedIn searches with automated, enriched lead lists.
firecrawl-market-research feeds the product roadmap with competitive intelligence — feature gaps, pricing opportunities, and positioning insights.

Research & Intelligence¶

firecrawl-deep-research replaces hours of manual competitive analysis. One command produces a cited research report.
firecrawl-seo-audit maintains corpusiq-docs' search visibility across 1,000+ pages.

Integration Pattern¶

All workflows can be chained into automated research pipelines:

#!/bin/bash
# Weekly competitive intelligence pipeline
# Save to: ~/.hermes/profiles/corpusiq/cron/weekly-competitive-intel.sh

WEEK=$(date +%Y-W%V)
OUTDIR=~/corpusiq-brain/research/$WEEK
mkdir -p $OUTDIR

# 1. Market landscape
firecrawl-market-research \
    --company "CorpusIQ" \
    --competitors "Glean, Notion AI, Mem, Tana, Coda AI" \
    > $OUTDIR/market-landscape.md

# 2. Deep dive on top competitor
firecrawl-deep-research \
    "Glean AI enterprise search features pricing 2026" \
    --depth 3 --sources 15 \
    > $OUTDIR/glean-deep-dive.md

# 3. SEO gap analysis
firecrawl-seo-audit https://corpusiq.io \
    --competitors "glean.com" \
    > $OUTDIR/seo-gap-analysis.md

# 4. Import to GBrain for semantic search
cd ~/corpusiq-brain && gbrain import research/$WEEK/ --no-embed

Troubleshooting¶

Symptom	Cause	Fix
`FIRECRAWL_API_KEY not set`	Missing API key	Sign up at firecrawl.dev, export key
Rate limited (429)	Free tier limits	Upgrade or add delays between calls
Empty lead results	Too restrictive filters	Broaden `--industry` or remove `--domains`
Research too shallow	Low `--depth`	Increase to 3-4 for comprehensive coverage
SEO audit timeout	Large site	Add `--pages 100` to limit crawl scope