Scan any company in 5 seconds. Tech stack, growth signals, conviction score, AI outreach.

Install · Commands · AI Setup · Library API

Install

npm install -g leadscan

Or run without installing:

npx leadscan analyze stripe.com

Commands

analyze <domain>

Full intelligence report on any company — tech stack, growth signals, contact info, and a 0–100 conviction score. No API key needed.

leadscan analyze stripe.com
leadscan analyze linear.app --json > linear.json

With AI analysis (set GROQ_API_KEY — free key here):

GROQ_API_KEY=your_key leadscan analyze notion.so

Adds: company summary, industry classification, target customer profile, and value proposition.

outreach <domain>

Generates a personalized cold email, LinkedIn message, and call opener for a specific role at the company.

leadscan outreach vercel.com --role "CTO" --product "CI/CD monitoring tool"
leadscan outreach hubspot.com --role "VP of Sales" --product "sales training platform"

| Flag | Required | Description | |---|---|---| | -r, --role <role> | Yes | Who you're targeting | | -p, --product <product> | Yes | What you're selling | | --json | No | Output raw JSON |

Without GROQ_API_KEY: basic template with company details substituted in.
With GROQ_API_KEY: every message references specific things found about the company.

batch <file>

Process a list of domains from a plain text file. Ranks by conviction score. Supports JSON export.

# targets.txt
stripe.com
linear.app
vercel.com
notion.so
figma.com

leadscan batch targets.txt --min-score 60 --output qualified.json

| Flag | Default | Description | |---|---|---| | --min-score <n> | 0 | Only include companies above this score | | --output <file> | — | Save results to JSON instead of printing | | --ai | false | Use AI analysis for each company |

Conviction Score

Every company gets a 0–100 conviction score built from four weighted signals:

| Signal | Weight | What it measures | |---|---|---| | Tech modernity | 25% | Modern stack (React, Vercel, Stripe, PostHog) vs. legacy (WordPress, Bootstrap) | | Growth signals | 35% | Actively hiring, funded, has pricing page, has a blog | | Market presence | 20% | LinkedIn, Twitter, GitHub, Crunchbase presence | | Contactability | 20% | Emails on the page, LinkedIn, hiring pages |

Score guide: 75+ = strong target · 50–74 = worth a look · <50 = low priority

Tech Stack Detection

Detects 40+ technologies across 6 categories from homepage HTML and script sources. Zero external API calls.

| Category | Technologies detected | |---|---| | Frontend | React, Next.js, Vue, Svelte, Remix, Nuxt, Tailwind, Alpine.js | | Backend | Node.js, Django, Rails, Laravel, WordPress, Shopify, Webflow, Framer | | Analytics | Google Analytics, Segment, Mixpanel, Amplitude, PostHog, Heap, Hotjar, Plausible | | Marketing | HubSpot, Intercom, Drift, Zendesk, Mailchimp, ActiveCampaign, Customer.io | | Infrastructure | Vercel, Cloudflare, AWS, Google Cloud, Fastly, Netlify, Firebase | | Payments | Stripe, Paddle, Chargebee, Recurly, Braintree, PayPal, Square |

AI Setup (optional)

Get a free API key at console.groq.com — no credit card required.

export GROQ_API_KEY=your_key_here
leadscan analyze linear.app

Uses Llama 3.3 70B via Groq. The tool works fully without a key — AI adds company summary, industry detection, and personalized outreach generation.

Use as a library

leadscan also exports a typed API for use in your own projects:

import { scrapeCompany, buildIntelReport, analyzeCompany, generateOutreach } from "leadscan";

// Scrape + score
const data   = await scrapeCompany("stripe.com");
const report = buildIntelReport(data);

console.log(report.convictionScore);     // 75
console.log(report.techStack.frontend);  // ["React", "Next.js"]
console.log(report.signals.isHiring);    // true
console.log(report.emails);             // ["[email protected]"]

// With AI (requires GROQ_API_KEY)
const ai = await analyzeCompany(data);
console.log(ai.industry);        // "Fintech"
console.log(ai.targetCustomer);  // "online businesses of all sizes"

// Generate outreach
const outreach = await generateOutreach(data, "Head of Engineering", "observability tool", ai);
console.log(outreach.email.subject);  // personalized subject line
console.log(outreach.linkedin);       // LinkedIn message

Full TypeScript types exported — see src/lib/types.ts for the complete interface.

Output formats

Every command supports --json for piping into other tools:

leadscan analyze stripe.com --json | jq '.convictionScore'
leadscan batch targets.txt --output results.json
cat results.json | jq 'sort_by(-.convictionScore) | .[0].domain'

Contributing

PRs welcome. The project structure:

src/
  cli.ts              # Commander entry point
  commands/
    analyze.ts        # analyze command
    outreach.ts       # outreach command
    batch.ts          # batch command
  lib/
    scraper.ts        # HTML fetching + parsing (Cheerio)
    scorer.ts         # 4-signal conviction scoring model
    ai.ts             # Groq AI integration (company analysis + outreach)
    display.ts        # Terminal output (chalk + cli-table3)
    types.ts          # All TypeScript interfaces

git clone https://github.com/nadellasripad11/leadscan
cd leadscan
npm install
npm run build
node dist/cli.js analyze stripe.com

License

MIT — Sripad Nadella