Lead Gen Pipeline

Automated lead generation pipeline with AI-powered lead scoring and personalized follow-up generation. Score leads 0-100 with reasoning, generate context-awa...

Security Scan

The skill's code, instructions, and required credential (OPENROUTER_API_KEY) are coherent with its stated purpose (lead scoring and follow-up generation) and do not request unrelated privileges.

Purpose & Capability

Name/description match the included Python scripts and SKILL.md. The only required credential is OPENROUTER_API_KEY which is needed to call the OpenRouter LLM API used by both scripts. No unrelated services, binaries, or config paths are requested.

Instruction Scope

SKILL.md instructs running the two included scripts with JSON input; the scripts only build JSON prompts and POST them to openrouter.ai. There are no instructions to read system files, secrets beyond OPENROUTER_API_KEY, or to exfiltrate data to unexpected endpoints. Example CRM integration lines reference external crm wrappers but are examples and not executed by the skill itself.

Install Mechanism

This is instruction-only / script-based with no install spec. Nothing is downloaded or installed by the skill, so there is no high-risk install mechanism.

Credentials

Only OPENROUTER_API_KEY is required (declared as primaryEnv). That is proportional: both scripts embed the key in Authorization headers to call openrouter.ai. No extra tokens, passwords, or unrelated env vars are requested.

Persistence & Privilege

The skill does not request always: true, does not modify other skills or system configuration, and is user-invocable. It runs only when the user runs the scripts.

Assessment

This skill is internally consistent, but note: using it will send lead data (names, company, context, behavioral signals, etc.) to OpenRouter (openrouter.ai). Before installing or running, ensure you have the right to send any personally identifiable or sensitive customer data to a third-party LLM. Use a dedicated OpenRouter API key with minimal permissions, test with scrubbed/example data first, and verify any CRM integration commands (the SKILL.md shows example paths to external crm wrappers that are not included). Also verify the homepage/author if provenance matters to you. If you need stricter data controls, consider running a local/private LLM or redacting PII before calling the API.

AI-powered lead generation pipeline. Score leads intelligently, generate personalized follow-ups, and manage your sales pipeline.

Quick Start

export OPENROUTER_API_KEY="your-key"

# Score a lead
python3 {baseDir}/scripts/lead_scorer.py '{"name":"Jane Smith","company":"Acme Corp","title":"VP Marketing","source":"webinar","actions":["downloaded whitepaper","visited pricing page 3x","opened 5 emails"]}'

# Generate follow-up
python3 {baseDir}/scripts/followup_generator.py '{"name":"Jane Smith","company":"Acme Corp","context":"Attended our AI webinar, downloaded whitepaper","stage":"warm","tone":"professional"}'

Lead Scoring

The AI scorer evaluates leads on multiple dimensions:

Factor	Weight	Description
Fit	30%	Does the lead match your ICP? (title, company size, industry)
Intent	30%	Behavioral signals (page visits, downloads, email engagement)
Engagement	20%	How actively are they interacting? (recency, frequency)
Source Quality	20%	Where did they come from? (referral > webinar > cold)

Score Ranges

• 80-100: 🔥 Hot — reach out immediately, high buying intent
• 60-79: 🟡 Warm — nurture with targeted content, book a call
• 40-59: 🟠 Cool — add to drip sequence, monitor engagement
• 0-39: 🔵 Cold — low priority, long-term nurture only

# Score with custom ICP
python3 {baseDir}/scripts/lead_scorer.py '{"name":"...","company":"...","icp":{"industries":["SaaS","fintech"],"minEmployees":50,"titles":["VP","Director","C-suite"]}}'

Follow-Up Generation

Generate personalized follow-up messages for any pipeline stage:

# Professional follow-up after demo
python3 {baseDir}/scripts/followup_generator.py '{
  "name": "Jane Smith",
  "company": "Acme Corp",
  "context": "Had a 30-min demo, interested in enterprise plan, concerned about onboarding time",
  "stage": "post-demo",
  "tone": "professional",
  "channel": "email"
}'

# Casual SMS check-in
python3 {baseDir}/scripts/followup_generator.py '{
  "name": "Mike",
  "context": "Met at conference, exchanged cards, talked about AI automation",
  "stage": "initial",
  "tone": "casual",
  "channel": "sms"
}'

# Urgent closing message
python3 {baseDir}/scripts/followup_generator.py '{
  "name": "Sarah Johnson",
  "company": "TechFlow",
  "context": "Proposal sent 5 days ago, no response, deal worth $25k, quarter ending",
  "stage": "closing",
  "tone": "urgent",
  "channel": "email"
}'

Supported Tones

• professional — formal business communication
• casual — friendly, conversational
• urgent — time-sensitive, action-oriented
• friendly — warm, relationship-focused
• consultative — expert advice framing

Supported Channels

• email — full email with subject line
• sms — short, punchy (< 160 chars)
• whatsapp — conversational, emoji-friendly
• linkedin — professional networking tone

Pipeline Stages

• initial — first contact / cold outreach
• warm — engaged but no meeting yet
• booked — meeting/demo scheduled
• post-demo — after initial call/demo
• proposal — proposal sent
• closing — negotiation / final decision
• revival — re-engaging cold/lost lead

Cold Outreach Templates

The AIDA Framework

1. Attention — Hook with relevant pain point
2. Interest — Show you understand their world
3. Desire — Paint the outcome
4. Action — Clear, low-friction CTA

Outreach Sequences

Day 1: Initial value-first email Day 3: Follow-up with case study / social proof Day 7: Different angle (video, voice note, meme) Day 14: Break-up email ("Should I close your file?")

Generate any of these:

python3 {baseDir}/scripts/followup_generator.py '{"name":"...","stage":"initial","sequence_step":1}'
python3 {baseDir}/scripts/followup_generator.py '{"name":"...","stage":"initial","sequence_step":4}'

CRM Integration Patterns

With GHL (GoHighLevel)

# 1. Score incoming lead
SCORE=$(python3 {baseDir}/scripts/lead_scorer.py '{"name":"...","source":"facebook_ad"}')

# 2. Create contact in GHL with score tag
python3 ../ghl-crm/{baseDir}/scripts/ghl_api.py contacts create '{"firstName":"...","tags":["score-85","hot-lead"]}'

# 3. Add to appropriate pipeline stage
python3 ../ghl-crm/{baseDir}/scripts/ghl_api.py opportunities create '{"pipelineId":"...","stageId":"hot-stage-id","contactId":"..."}'

# 4. Generate and send follow-up
MSG=$(python3 {baseDir}/scripts/followup_generator.py '{"name":"...","stage":"warm","channel":"sms"}')
python3 ../ghl-crm/{baseDir}/scripts/ghl_api.py conversations send-sms <contactId> "$MSG"

With Any CRM

The scripts output JSON — pipe into any CRM API wrapper. Lead scores include reasoning that can be stored as CRM notes.

Response Handling

When a lead replies, re-score with updated context:

python3 {baseDir}/scripts/lead_scorer.py '{"name":"Jane","company":"Acme","actions":["replied to email","asked about pricing","requested demo"]}'

Then generate contextual response:

python3 {baseDir}/scripts/followup_generator.py '{"name":"Jane","context":"She asked about pricing and wants a demo","stage":"warm","tone":"professional"}'

Credits

Built by M. Abidi | agxntsix.ai YouTube | GitHub Part of the AgxntSix Skill Suite for OpenClaw agents.

📅 Need help setting up OpenClaw for your business? Book a free consultation