What it does
Open-source AI marketing automation skills for growth experiments, sales pipeline, content ops, outbound, SEO, and finance automation
Skills ZICQ category:Agent Workflows ai-marketing-skills-automation
Open-source AI marketing automation skills for growth experiments, sales pipeline, content ops, outbound, SEO, and finance automation
Official URL:skills.sh
Intro in this page language first. The official description stays in its original wording; we do not rewrite SKILL.md.
Open-source AI marketing automation skills for growth experiments, sales pipeline, content ops, outbound, SEO, and finance automation
The official description does not include a separate “Use when”. Per the spec, agents activate this skill when the task matches keywords in that description.
Per Agent Skills progressive disclosure: name and description load at startup (~100 tokens); the full SKILL.md body loads when the skill activates; scripts/, references/, and assets/ load only as needed. This file's sections: AI Marketing Skills Automation; What It Does; Installation; Clone the repository; Navigate to a specific skill category; Install dependencies for that category.
File analysis: besides SKILL.md, the body references references/cold_email_template.md. Those resources load on demand. The body is about 560 lines, above the spec’s 500-line guidance; agents still load it in full on activation.
AI Marketing Skills AutomationWhat It DoesInstallationClone the repositoryNavigate to a specific skill categoryInstall dependencies for that categorySet up environment variablesConfigurationCommon environment variables across skillsGrowth Engine specificSales Pipeline specificSEO Ops specific
Source category:skills.sh agent-skill
nameai-marketing-skills-automationdescriptionreferences/cold_email_template.mdThese paths are extracted from the text. Check the upstream package to verify the files exist.
Invocation syntax and available tools depend on your Agent client. Client integration guide ↗
Choose the target agent and installation scope, keep referenced package files, then verify the skill appears in the client's catalog.
This skill references supporting files. Retrieve the complete directory from the source; copying SKILL.md alone may leave missing dependencies.
Copy these instructions to a compatible agent and confirm the target directory matches your client.
Install the agent skill "ai-marketing-skills-automation" into my project. The full SKILL.md and official description are at https://zicq.com/en/skills/skl-7072cdacf6f9203e-Ai-Marketing-Skills-Automation.html Save it as .cursor/skills/ai-marketing-skills-automation/SKILL.md or .claude/skills/ai-marketing-skills-automation/SKILL.md and keep the frontmatter name and description exactly as-is. This skill also ships scripts/, references/, or assets/ — fetch the whole folder from https://github.com/reason-machines/marketing-skills instead of creating only a SKILL.md.
Requires Node.js and npx. First inspect the repository's skill list to confirm the name.
npx skills add 'https://github.com/reason-machines/marketing-skills' --list
npx skills add 'https://github.com/reason-machines/marketing-skills' --skill 'ai-marketing-skills-automation'
The CLI lets you choose the agent interactively. The default scope is the project; use -g for user scope. Confirm package availability with the discovery command, then use npx skills list to inspect installed skills.
name: ai-marketing-skills-automation description: Open-source AI marketing automation skills for growth experiments, sales pipeline, content ops, outbound, SEO, and finance automation triggers:
Skill by ara.so — Marketing Skills collection.
This project provides battle-tested marketing automation workflows — not prompts, but complete Python scripts with scoring algorithms, expert panels, and automation pipelines. Built for Claude Code and other AI coding agents to execute real marketing operations.
AI Marketing Skills gives you 15+ categories of marketing automation:
# Clone the repository
git clone https://github.com/ericosiu/ai-marketing-skills.git
cd ai-marketing-skills
# Navigate to a specific skill category
cd growth-engine # or sales-pipeline, content-ops, etc.
# Install dependencies for that category
pip install -r requirements.txt
# Set up environment variables
cp .env.example .env
Each category uses a .env file for API keys and configuration:
# Common environment variables across skills
ANTHROPIC_API_KEY=your_anthropic_key_here
OPENAI_API_KEY=your_openai_key_here
# Growth Engine specific
GOOGLE_ANALYTICS_KEY=your_ga_key
LINKEDIN_API_KEY=your_linkedin_key
# Sales Pipeline specific
RB2B_API_KEY=your_rb2b_key
INSTANTLY_API_KEY=your_instantly_key
APOLLO_API_KEY=your_apollo_key
# SEO Ops specific
GOOGLE_SEARCH_CONSOLE_CREDENTIALS=path/to/credentials.json
# Revenue Intelligence specific
GONG_API_KEY=your_gong_key
SALESFORCE_API_KEY=your_salesforce_key
Run autonomous marketing experiments with statistical rigor.
from experiment_engine import ExperimentEngine
# Initialize the engine
engine = ExperimentEngine(
api_key=os.getenv("ANTHROPIC_API_KEY"),
data_source="google_analytics"
)
# Create an experiment
experiment = engine.create_experiment(
hypothesis="Thread posts get 2x engagement vs single posts",
variable="format",
variants=["thread", "single"],
metric="impressions",
duration_days=14,
traffic_split=0.5
)
# Run the experiment
results = engine.run_experiment(experiment.id)
# Get statistical significance
analysis = engine.analyze_results(
experiment.id,
confidence_level=0.95,
test_method="mann_whitney"
)
print(f"Winner: {analysis['winner']}")
print(f"P-value: {analysis['p_value']}")
print(f"Confidence: {analysis['confidence_interval']}")
from pacing_alert import PacingMonitor
monitor = PacingMonitor(
budget_monthly=10000,
platform="linkedin"
)
# Check daily pacing
alert = monitor.check_pacing(
spend_to_date=3500,
days_elapsed=8,
days_in_month=30
)
if alert['status'] == 'overpacing':
print(f"Alert: Overpacing by {alert['variance_percent']}%")
print(f"Recommended daily spend: ${alert['recommended_daily']}")
# Create experiment
python experiment-engine.py create \
--hypothesis "Carousel posts outperform static images" \
--variable post_type \
--variants '["carousel", "static"]' \
--metric engagement_rate \
--duration 14
# Check pacing
python pacing-alert.py check \
--budget 10000 \
--spend 3500 \
--days-elapsed 8
# Generate weekly scorecard
python autogrowth-weekly-scorecard.py generate \
--start-date 2026-05-01 \
--end-date 2026-05-07
Turn anonymous website visitors into qualified pipeline.
from rb2b_instantly_router import RB2BRouter
router = RB2BRouter(
rb2b_key=os.getenv("RB2B_API_KEY"),
instantly_key=os.getenv("INSTANTLY_API_KEY")
)
# Fetch website visitors from RB2B
visitors = router.fetch_visitors(
lookback_hours=24,
min_intent_score=7
)
# Route to Instantly campaigns
for visitor in visitors:
# Score and enrich
enriched = router.enrich_visitor(visitor)
# Route based on criteria
if enriched['seniority'] in ['C-Level', 'VP', 'Director']:
router.add_to_campaign(
email=enriched['email'],
campaign_id="high-intent-vp",
personalization={
'company': enriched['company'],
'trigger': enriched['page_visited']
}
)
from deal_resurrector import DealResurrector
resurrector = DealResurrector(
crm_api_key=os.getenv("SALESFORCE_API_KEY")
)
# Find stale deals with departed champions
stale_deals = resurrector.find_stale_deals(
days_inactive=90,
min_deal_value=10000
)
# Track champions to new companies
for deal in stale_deals:
champion_moves = resurrector.track_champion(
champion_email=deal['primary_contact'],
linkedin_api_key=os.getenv("LINKEDIN_API_KEY")
)
if champion_moves['new_company']:
resurrector.create_new_opportunity(
company=champion_moves['new_company'],
contact=champion_moves['new_email'],
context=deal['previous_context']
)
from icp_learning_analyzer import ICPLearner
learner = ICPLearner()
# Analyze win/loss patterns
deals = learner.fetch_closed_deals(months_back=12)
patterns = learner.analyze_patterns(deals)
# Update ICP definition
new_icp = learner.update_icp(
current_icp="""
Company size: 50-500 employees
Industry: SaaS, E-commerce
Tech stack: React, Python
""",
win_loss_data=patterns
)
print(new_icp)
Ship content that scores 90+ every time.
from expert_panel import ExpertPanel
panel = ExpertPanel(
api_key=os.getenv("ANTHROPIC_API_KEY")
)
# Load expert personas
panel.load_experts([
'experts/seo_expert.json',
'experts/conversion_expert.json',
'experts/content_strategist.json'
])
# Score content
content = """
Your blog post content here...
"""
scores = panel.score_content(
content=content,
rubric='scoring-rubrics/blog_post.json',
min_score=90
)
# Recursive improvement
while scores['average'] < 90:
feedback = panel.get_improvement_suggestions(scores)
content = panel.improve_content(content, feedback)
scores = panel.score_content(content, 'scoring-rubrics/blog_post.json')
print(f"Final score: {scores['average']}")
print(f"Expert breakdown: {scores['by_expert']}")
# CLI quality gate
python quality-gate.py check \
--file blog-post.md \
--rubric scoring-rubrics/blog_post.json \
--min-score 90 \
--experts seo conversion content-strategy
ICP to inbox automation.
from cold_outbound_optimizer import OutboundEngine
engine = OutboundEngine(
apollo_key=os.getenv("APOLLO_API_KEY"),
instantly_key=os.getenv("INSTANTLY_API_KEY")
)
# Define ICP
icp = {
'titles': ['VP Marketing', 'CMO', 'Head of Growth'],
'company_size': [50, 500],
'industries': ['SaaS', 'E-commerce'],
'technologies': ['HubSpot', 'Salesforce']
}
# Build lead list
leads = engine.build_lead_list(
icp=icp,
limit=1000,
exclude_domains=['competitor1.com', 'competitor2.com']
)
# Generate personalized emails
for lead in leads:
email = engine.generate_email(
lead=lead,
template='references/cold_email_template.md',
personalization_depth='high'
)
engine.add_to_sequence(
email=lead['email'],
campaign='q2-outbound',
message=email
)
Find keywords your competitors missed.
from content_attack_brief import SEOBrief
brief = SEOBrief(
gsc_credentials=os.getenv("GOOGLE_SEARCH_CONSOLE_CREDENTIALS")
)
# Analyze content gaps
gaps = brief.find_content_gaps(
target_domain='yoursite.com',
competitor_domains=['competitor1.com', 'competitor2.com'],
topic='marketing automation'
)
# Generate brief
content_brief = brief.generate_brief(
keyword=gaps[0]['keyword'],
search_intent=gaps[0]['intent'],
top_ranking_urls=gaps[0]['serp_results']
)
print(content_brief)
# CLI GSC optimization
python gsc_client.py analyze \
--domain yoursite.com \
--lookback-days 90 \
--min-impressions 1000 \
--position-range 11-20
AI CFO for cost analysis.
from cfo_briefing import FinanceAnalyzer
analyzer = FinanceAnalyzer()
# Upload financial data
analyzer.load_data(
expenses='data/expenses_q1.csv',
revenue='data/revenue_q1.csv'
)
# Generate CFO briefing
briefing = analyzer.generate_briefing(
focus_areas=['hidden_costs', 'vendor_optimization', 'budget_variance']
)
# Get cost-saving recommendations
recommendations = analyzer.find_savings_opportunities(
min_impact=5000 # Minimum $5k annual savings
)
print(briefing)
for rec in recommendations:
print(f"{rec['category']}: Save ${rec['annual_savings']:,.0f}")
Sales call insights and attribution.
from gong_insight_pipeline import GongAnalyzer
analyzer = GongAnalyzer(
gong_api_key=os.getenv("GONG_API_KEY")
)
# Fetch recent calls
calls = analyzer.fetch_calls(
date_range='last_7_days',
min_duration_minutes=20
)
# Extract insights
for call in calls:
insights = analyzer.extract_insights(call['id'])
# Key patterns
print(f"Objections: {insights['objections']}")
print(f"Competitor mentions: {insights['competitors']}")
print(f"Next steps: {insights['next_steps']}")
# Update CRM
analyzer.sync_to_crm(
call_id=call['id'],
insights=insights,
crm='salesforce'
)
# All scripts include retry logic with exponential backoff
from utils import retry_with_backoff
@retry_with_backoff(max_retries=5, base_delay=2)
def api_call():
return client.make_request()
# Scan for sensitive data before commits
python3 security/sanitizer.py --scan --dir . --recursive
# Install pre-commit hook
cp security/pre-commit-hook.sh .git/hooks/pre-commit
chmod +x .git/hooks/pre-commit
# Each category has isolated dependencies
cd growth-engine
pip install --upgrade -r requirements.txt
# If conflicts, use virtual environment
python -m venv venv
source venv/bin/activate # or venv\Scripts\activate on Windows
pip install -r requirements.txt
All scripts sanitize PII by default:
from security.sanitizer import sanitize_output
# Automatically removes emails, phone numbers, API keys
safe_data = sanitize_output(raw_data)
# Example: SEO → Content → Quality Gate → Publish
from content_attack_brief import SEOBrief
from expert_panel import ExpertPanel
# 1. Get SEO brief
brief = SEOBrief().generate_brief(keyword='ai marketing automation')
# 2. Generate content
content = generate_from_brief(brief) # Your content generation
# 3. Score with expert panel
panel = ExpertPanel()
scores = panel.score_content(content, min_score=90)
# 4. Publish if passed
if scores['average'] >= 90:
publish_to_cms(content)
# View local usage stats
python3 telemetry/telemetry_report.py
# Check for updates
python3 telemetry/version_check.py
# Opt out of remote telemetry (local logging still works)
export AI_MARKETING_SKILLS_TELEMETRY=false
ai-marketing-skills/
├── growth-engine/ # Experiments, pacing, scorecards
├── sales-pipeline/ # RB2B, deal resurrector, ICP learner
├── content-ops/ # Expert panel, quality gates
├── outbound-engine/ # Cold email automation
├── seo-ops/ # Content gaps, GSC analysis
├── finance-ops/ # CFO briefings, cost analysis
├── revenue-intelligence/ # Gong insights, attribution
├── conversion-ops/ # CRO audits, lead magnets
├── podcast-ops/ # Episode → content pipeline
├── sales-playbook/ # Value pricing frameworks
├── autoresearch/ # Evolutionary content optimization
├── deck-generator/ # AI slide decks
├── yt-competitive-analysis/ # YouTube outlier detection
└── x-longform-post/ # Human-sounding X posts
Each category contains:
SKILL.md — Category-specific skill documentationscripts/ — Python automation scriptsrequirements.txt — Dependencies.env.example — Configuration templateREADME.md — Category guideAgent Workflows
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that exte…
Agent Workflows
Use the ClawdHub CLI to search, install, update, and publish agent skills from clawdhub.com. Use when you need to fetch new skills on the fl…
Agent Workflows
Orchestrate multi-agent teams with defined roles, task lifecycles, handoff protocols, and review workflows. Use when: (1) Setting up a team …
Agent Workflows
Spec-first, TDD, subagent-driven software development workflow. Use when: (1) building any new feature or app — triggers brainstorm → plan →…