功能描述

Automated lead generation pipeline that finds local businesses with weak/no websites, AI-generates custom demo sites, deploys to Vercel, and runs a 5-email c...

使用说明 (SKILL.md)

Lead Gen Website Pipeline

Name: Lead Gen Website Pipeline
Author: razzledazzlei

Automated system that finds local businesses with weak or no websites, AI-generates a custom demo website for each, deploys it to Vercel, and sends a personalized cold outreach email sequence — all without human intervention after initial setup.

What It Does

Discovers leads via Google Places API for configurable niches and cities
Scores leads 0-100 based on website quality, reviews, phone availability
Queues leads in a Google Sheet for human review/approval
Generates demo websites via a 5-stage AI pipeline (research → spec → build → QA → validate)
Deploys to Vercel automatically with daily rate limiting
Sends cold outreach via AgentMail with a 5-email drip sequence (Day 0, 3, 7, 14, 30)
Notifies you on Discord with pitch angles, demo URL, and visual QA scores

Pipeline Architecture

Google Places API → Score Leads → Google Sheet (you approve)
       ↓
  Approved lead → Scrape website → Fetch Google reviews
       ↓
  AI Research → AI Spec (copy) → AI Build (HTML) → AI QA → Validate
       ↓
  Deploy to Vercel → Update Sheet → Discord notification
       ↓
  Generate outreach email → AgentMail sends → Drip follow-ups

Requirements

API Keys (set in your .env)

Key	Service	Purpose	Cost
`GOOGLE_PLACES_API_KEY`	Google Cloud	Lead discovery + photos	~$0-20/mo
`GOOGLE_API_KEY`	Google AI Studio	Gemini for research/spec/QA	Free tier
`ANTHROPIC_API_KEY`	Anthropic	Claude for spec generation	Pay per use
`OPENAI_API_KEY`	OpenAI	GPT for HTML building	Pay per use
`VERCEL_TOKEN`	Vercel	Demo site deployment	Free tier (100/day)
`AGENTMAIL_API_KEY`	AgentMail	Cold outreach emails	Free tier
`GOOGLE_SHEET_ID`	Google Sheets	Lead tracking	Free

Google Sheet Setup

Status values: review, filtered, approved, processing, demo_ready, failed, contacted

Share the sheet with a GCP service account for API access.

Infrastructure

Node.js 22+ on the host machine
Google Cloud service account JSON key for Sheets API
Vercel account (free Hobby tier works)
AgentMail account with at least one inbox
Discord channel for notifications (optional but recommended)

Setup Instructions

1. Clone and install

git clone https://github.com/RazzleDazzleI/lead-gen-pipeline.git
cd lead-gen-pipeline
npm install

2. Configure environment

Copy .env.example to .env and fill in your API keys.

3. Configure your niches and city

Edit scripts/niche-defaults.json to customize:

Target niches (landscaping, cleaning, roofing, etc.)
Color palettes per niche
Common services per niche
Trust items per niche

The pipeline supports 15 niches out of the box: landscaping, auto detailer, cleaning service, handyman, roofing, pressure washing, plumbing, electrician, HVAC, painting, fencing, tree service, junk removal, pest control

4. Set up the poller

The poller watches your Google Sheet for approved leads and triggers the pipeline:

# Run once to test
node scripts/poll-approved-leads.js

# Run continuously (every 5 minutes)
node scripts/run-poller.js 5

For production, set up as a systemd service or OpenClaw cron job.

5. Set up outreach emails

# Dry run first — see what would be sent
node scripts/send-outreach.js --dry-run

# Send pending initial emails + follow-ups
node scripts/send-outreach.js

# Only send scheduled follow-ups
node scripts/send-outreach.js --followups

Set up as an OpenClaw cron job to run 3x/day:

openclaw cron add --name "outreach-sender" \
  --cron "0 9,15,21 * * *" --tz "America/Chicago" \
  --session isolated --announce \
  --message "Run: cd /path/to/project && node scripts/send-outreach.js"

Customization

Target City

Change the Google Places API search query in the n8n workflow or lead discovery script to target your city instead of Omaha.

Pricing Tiers

The business plan template includes 3 tiers:

Starter ($997 + $49/mo) — Single-page landing site
Professional ($1,997 + $99/mo) — Multi-page with gallery, analytics
Premium ($2,997 + $149/mo) — Blog, booking, live chat, priority support

Email Templates

The drip sequence (in send-outreach.js) follows the proven pattern:

Day 0: "I built you a website — take a look" (demo link)
Day 3: Social proof — "businesses like yours saw 30-40% more calls"
Day 7: Urgency — "your preview expires soon"
Day 14: Last chance — "taking it down today"
Day 30: Breakup — "closing the loop, door is always open"

Edit the templates in send-outreach.js → getDripEmail() function.

Design Blueprints

Add niche-specific design guidance in templates/designs/\x3Cniche>.md. The build stage loads these automatically to produce better-looking sites per vertical.

Pipeline Stages Deep Dive

Stage 1: Research

Consolidates lead data (name, phone, address, website, reviews) into a structured JSON summary. Uses scraped website data + Google Places API reviews.

Stage 2: Spec

Generates all website copy — hero section, services, about, reviews, service area, footer. No HTML yet, just structured content.

Stage 3: Build

Converts the spec JSON into a single-file responsive HTML page. Injects real photos (scraped or Google Places fallback), logo, niche-specific color palette.

Stage 4: QA

AI reviews the HTML for 13 quality checks: robots meta tag, demo banner, phone numbers, form action, footer attribution, broken images, alt text, mobile hamburger menu, and more.

Stage 5: Validate

Programmatic validation confirms all critical elements are present. Non-blocking — logs warnings but doesn't block deployment.

Stage 6: Deploy

POSTs the final HTML to Vercel API. Tracks daily deploy count (max 90/day to stay under free tier limit). Returns live URL.

Stage 7: Outreach

Generates a personalized cold email with pitch angles extracted from the lead data (no website, Gmail address, strong reviews, years in business).

Output Per Lead

Each processed lead produces a run folder with ~19 artifacts:

intake.json — Enriched lead data
research.json — Structured summary
spec.json — All website copy
final.html — Production-ready HTML
evaluation.json — Deploy URL + metadata
outreach-email.json — Cold email + follow-up
screenshot-*.png — Mobile/tablet/desktop captures
visual-qa.json — Gemini Vision quality scores
status.json — Run tracking

Costs

Per-Lead Cost

Google Places API: ~$0.03-0.05/lead
AI generation (Gemini + Claude): ~$0.10-0.50/lead
Vercel hosting: $0 (free tier)
AgentMail: $0 (free tier)
Total: ~$0.15-0.55 per lead

Monthly Operating Cost (at 10-20 leads/day)

Google APIs: $5-20/mo
AI models: $30-150/mo (less with Claude Max plan)
Everything else: $0
Total: $35-170/mo

Revenue Potential (at 2% response rate)

50 emails/week → 1 conversation/week → 1 client/month
1 client = $997+ one-time + $49+/mo recurring
Year 1 from one client/month: ~$12,000+ one-time + $3,500+ recurring

Selling This System

This pipeline is designed to be sold as a done-for-you service:

Model	Setup Fee	Monthly	Your Time
Done-for-you	$2,000-$10,000	$200-$1,000/mo	4-8 hours setup
Pre-configured package	$500-$3,000	$100-$500/mo	1-2 hours per client
Productized service	—	$1,500/mo flat	Template + light customization

The pipeline documents itself. Each run folder is a complete audit trail. The niche configs are templates. Second client takes 1/4 the time of the first.

Credits

Built by Ryan Romero (Romero Automation) with Larry (OpenClaw agent). Pipeline architecture, spec generation, QA validation, and outreach automation — all designed for local service businesses in any US city.

安全使用建议

Proceed with caution. Before cloning or running anything: 1) Review the referenced GitHub repository code yourself (npm packages, scripts/send-outreach.js, poller scripts) — do not run npm install blindly. 2) Expect to supply more secrets than the registry lists (Google Sheet ID, GCP service-account JSON, optional Anthropic/OpenAI keys, AgentMail inbox). Treat the GCP JSON and API tokens as sensitive; store them securely and use least-privilege service accounts. 3) Test in an isolated VM/container and run the pipeline in dry-run mode first (the README describes a --dry-run for outreach). 4) Be aware this automates scraping and cold emails (possible legal, spam, or terms-of-service issues); confirm this behavior is acceptable for your use case and compliant with local law and service TOS. 5) Consider restricting network/access (scoped service account, restricted tokens), rotate tokens after testing, and audit outgoing emails and deployed sites frequently. 6) Ask the publisher to update metadata to declare all required env vars and config paths (GOOGLE_SHEET_ID, GCP service-account path, optional LLM keys) so the risk surface is explicit.

功能分析

Type: OpenClaw Skill Name: lead-gen-website-pipeline Version: 1.0.1 The lead-gen-website-pipeline skill bundle automates business discovery, website generation, and cold outreach. It requires high-privilege credentials (VERCEL_TOKEN, GOOGLE_API_KEY, AGENTMAIL_API_KEY) and performs automated network actions including site deployments and email campaigns. While the instructions in SKILL.md and reference files (references/env-example.md) are consistent with the stated purpose, the high-risk nature of the required permissions and the automated execution of external scripts (referenced via GitHub) warrant a suspicious classification. No direct evidence of malicious intent or prompt injection was identified in the provided documentation.

能力评估

ℹ Purpose & Capability

The declared purpose (find local businesses, build demos, deploy to Vercel, send outreach) aligns with the environment variables and binaries that appear in the SKILL.md (Google APIs, Vercel token, AgentMail, node). However the SKILL.md lists additional API keys and configuration (GOOGLE_SHEET_ID, ANTHROPIC_API_KEY, OPENAI_API_KEY, service account JSON path, etc.) that are not included in the registry 'requires.env' nor in 'required config paths'. That omission is an incoherence: the skill will in practice need more secrets/config than the metadata declares.

⚠ Instruction Scope

The SKILL.md instructs the user/agent to clone an external GitHub repo and run npm install and long-running poller/send scripts, to save a GCP service account JSON to a specific path (~/.openclaw/workspace/gcp-service-account.json), and to run cron/systemd services that automate scraping, deployments, and outbound emails. These instructions request access to sensitive credentials and persistent execution and also reference environment variables and files not declared in the registry metadata. The instructions also direct scraping and bulk email sending (potential legal/ethical risk) and will cause external code to be fetched and executed on the host.

⚠ Install Mechanism

The skill is instruction-only (no install spec), but SKILL.md tells you to git clone a GitHub repo and run npm install. That means installing and running third-party code from the repo at runtime; while GitHub is a common source, this is effectively an unreviewed download-and-execute flow. The registry metadata did not include an install spec or packaged code, so a manual review of the repo is required before running.

⚠ Credentials

The registry 'requires.env' lists four variables, but the SKILL.md and reference files require many more (GOOGLE_SHEET_ID, ANTHROPIC_API_KEY, OPENAI_API_KEY, AGENTMAIL_INBOX, VERCEL_PROJECT_NAME, FORM_EMAIL, BRAND_DOMAIN, and a GCP service-account JSON at a fixed path). Service account JSON is effectively a full GCP credential and is not declared in 'required config paths'. Requiring broad credentials (Google APIs + service account + Vercel token + AgentMail key + optional LLM keys) is proportionate to the pipeline's functionality but the mismatch/undisclosed items are a red flag: the metadata understates the actual credentials that will be needed and accessible.

ℹ Persistence & Privilege

The skill is not force-installed (always:false) and does not declare autonomous invocation restrictions, but the pipeline instructions explicitly create persistent pollers/cron jobs and long-running processes that will automatically discover leads, deploy sites, and send drip emails. Autonomous, persistent email-sending and deployment functionality increases blast radius if secrets are compromised, so users should treat the created services as high-privilege and run them in an isolated environment. The skill itself does not request 'always:true', but its recommended setup yields persistent privileged activity.

版本历史

v1.0.1

- Updated documentation with the new SKILL.md, providing a thorough overview of setup, pipeline stages, and customization options. - Clarified API key requirements, email drip sequence, and infrastructure needs. - Added detailed Google Sheet column requirements and statuses for lead tracking. - Expanded customization and design blueprint instructions for different business niches. - Included transparent cost breakdowns and revenue projections.

元数据

Slug lead-gen-website-pipeline

版本 1.0.1

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题