AI Marketing Skills
Open-source Claude Code skills for marketing and sales teams. Built by the team at Single Brain — battle-tested on real pipelines generating millions in revenue.
These aren't prompts. They're complete workflows — scripts, scoring algorithms, expert panels, and automation pipelines you can plug into Claude Code (or any AI coding agent) and run today.
🗂️ Skills
| Category |
What It Does |
Key Skills |
| Growth Engine |
Autonomous marketing experiments that run, measure, and optimize themselves |
Experiment Engine, Pacing Alerts, Weekly Scorecard |
| Sales Pipeline |
Turn anonymous website visitors into qualified pipeline |
RB2B Router, Deal Resurrector, Trigger Prospector, ICP Learner |
| Content Ops |
Ship content that scores 90+ every time |
Expert Panel, Quality Gate, Editorial Brain, Quote Miner |
| Growth Signal Charts |
Turn original content and measured audience signals into cited social charts |
Content Spikes, Recent Research, Desk Edition, PNG/PDF Exports |
| Content OS Portable Starter |
Bootstrap a vendor-neutral, draft-first content control loop |
Config Schema, Approval Gates, Connector Matrix, Readback Templates |
| Personal Strategic Signal Intelligence |
Turn private reading, bookmarks, notes, and applied work into source-grounded decision intelligence |
Attention Drift, Decision Court, Bookmark-to-Build, Content Negative-Space |
| Outbound Engine |
ICP definition to emails in inbox — fully automated |
Cold Outbound Optimizer, Lead Pipeline, Competitive Monitor |
| SEO Ops |
Find the keywords your competitors missed |
Content Attack Briefs, GSC Optimizer, Trend Scout |
| Finance Ops |
Your AI CFO that finds hidden costs in 30 minutes |
CFO Briefing, Cost Estimate, Scenario Modeler |
| Revenue Intelligence |
Prove content ROI and turn sales calls into strategy |
Gong Insight Pipeline, Revenue Attribution, Client Report Generator |
| Conversion Ops |
Score any landing page and turn survey data into lead magnets |
CRO Audit, Survey-to-Lead-Magnet Engine |
| Podcast Ops |
One episode → 20+ content pieces across every platform |
Podcast-to-Everything Pipeline, Content Calendar |
| Team Ops |
Ruthless performance audits and meeting intelligence |
Elon Algorithm, Meeting-to-Action Extractor |
| Sidebar Prioritizer |
Reset Codex sidebar priorities and label tasks by work category and inactivity |
P0–P4 Priorities, B7 Age Markers, Follow-up Evidence, Verified Labels |
| Sales Playbook |
Value-based pricing framework that turns $10K deals into $100K deals |
Pre-Call Briefing, Tiered Packager, Call Analyzer, Pattern Library |
| Autoresearch |
Karpathy-inspired optimization loops for conversion content — 50+ variants, expert scoring, evolved winners |
Variant Generator, Expert Panel Scorer, Evolution Engine |
| Deck Generator |
AI-generated slide decks with consistent visual styles in minutes |
Image Generator, Google Slides Builder, Style Presets |
| YT Competitive Analysis |
Find outlier videos and packaging patterns across any YouTube channels |
Outlier Detector, Title Pattern Extractor, Channel Benchmarker |
| YouTube Packaging |
Create, render, revise, and learn from evidence-backed title-thumbnail packages |
Channel Profiles, Package Scoring, Visual QA, Performance Learning |
| Video Content Engine |
Diagnose any video and route it into the strongest justified content portfolio |
Long-form Optimizer, Shorts Engine, Tutorial, Case Study, Paid Creative |
| Video Analysis |
Read transcripts first and inspect video silently when visuals or timing matter |
Transcript Review, Demo Verification, Clip Boundaries |
| Agentic Video Understanding |
Goal-directed moment extraction from long video/audio via Gemini (cheaper than fixed-FPS ingest) |
Moment Finder, Hook Hunt, Call Mining, Loom Bar Evidence |
| Show-and-Tell Video Slate |
Turn real builds and first-party proof into ranked 15-minute video episodes before production |
Evidence Inventory, Episode Scoring, Packaging Gates, Shoot Order |
| Shortform Production |
Produce readable shorts and deliver approved clips through the API |
V5 Style, Hook Previews, Editorial Review, Metricool Delivery |
| Shortform Idea Grill |
Interview, score, and rank proof-led short-form video ideas |
Five-Second Hooks, Three-Second Scores, Talking Bullets, CTA Mapping |
| Shortform Format Library |
Maintain visual format references and match content ideas to recording outlines |
Content Spikes, Creator Discovery, Freshness, Edit Briefs, Lead-Magnet Handoff |
| Net-New Video Editor |
Turn fresh talking-head recordings into reversible, review-ready vertical drafts |
JSON Edit Plans, FFmpeg Rendering, Captions, Hook Cards, Visual QA |
| X Long-Form + Humanizer |
Write X articles that sound human — with a 24-pattern AI slop detector |
Post Writer, Humanizer Checklist, ASCII Diagram Builder |
🚀 Quick Start
Each skill category has its own README with setup instructions. The general pattern:
# 1. Clone the repo
git clone https://github.com/ericosiu/ai-marketing-skills.git
cd ai-marketing-skills
# 2. Pick a category
cd growth-engine # or sales-pipeline, content-ops, etc.
# 3. Install dependencies
pip install -r requirements.txt
# 4. Set up environment variables
cp .env.example .env
# Edit .env with your API keys
# 5. Run
python experiment-engine.py create \
--hypothesis "Thread posts get 2x engagement vs single posts" \
--variable format \
--variants '["thread", "single"]' \
--metric impressions
🧠 How These Work with Claude Code
Every category includes a SKILL.md file. Copy the complete skill directory when it also contains scripts/, references/, agents/, or other bundled files. Copying only SKILL.md will leave those workflows incomplete.
# In your project directory
cp ai-marketing-skills/growth-engine/SKILL.md .claude/skills/growth-engine.md
For a bundled skill such as YouTube Packaging, install the full directory:
cp -R ai-marketing-skills/packaging-youtube-thumbnails \
.claude/skills/packaging-youtube-thumbnails
Other agent harnesses should import the same complete directory and use its SKILL.md as the entry point. OpenAI-compatible harnesses can also read agents/openai.yaml for display metadata and the default prompt.
Then ask Claude Code: "Run an experiment testing carousel vs. static posts on LinkedIn" — it handles the rest.
📊 What Makes These Different
These aren't toy demos. Each skill was built to run real business operations:
- Growth Engine uses bootstrap confidence intervals and Mann-Whitney U tests — real statistics, not vibes
- Deal Resurrector has three intelligence layers including "follow the champion" — tracking departed contacts to their new companies
- ICP Learner rewrites your ideal customer profile based on actual win/loss data — your targeting improves automatically
- Expert Panel recursively scores content with domain-specific expert personas until quality hits 90+
- RB2B Router does intent scoring, seniority-based company dedup, and agency classification before routing to outbound sequences
📁 Repository Structure
ai-marketing-skills/
├── README.md ← You are here
├── growth-engine/ ← Autonomous experiments
│ ├── SKILL.md
│ ├── experiment-engine.py
│ ├── pacing-alert.py
│ ├── autogrowth-weekly-scorecard.py
│ └── ...
├── sales-pipeline/ ← Visitor → pipeline automation
│ ├── SKILL.md
│ ├── rb2b_instantly_router.py
│ ├── deal_resurrector.py
│ ├── trigger_prospector.py
│ ├── icp_learning_analyzer.py
│ └── ...
├── content-ops/ ← Quality scoring & production
│ ├── SKILL.md
│ ├── scripts/
│ ├── experts/ ← 9 expert panel definitions
│ ├── scoring-rubrics/ ← 5 scoring rubric templates
│ └── ...
├── personal-strategic-signal-intelligence/ ← Private signals → decision intelligence
│ ├── SKILL.md
│ └── README.md
├── outbound-engine/ ← Cold outbound automation
│ ├── SKILL.md
│ ├── scripts/
│ ├── references/ ← ICP template, copy rules
│ └── ...
├── seo-ops/ ← SEO intelligence
│ ├── SKILL.md
│ ├── content_attack_brief.py
│ ├── gsc_client.py
│ ├── trend_scout.py
│ └── ...
├── finance-ops/ ← Financial analysis
│ ├── SKILL.md
│ ├── scripts/
│ ├── references/ ← Metrics, rates, ROI models
│ └── ...
├── revenue-intelligence/ ← Sales call insights + attribution
│ ├── SKILL.md
│ ├── gong_insight_pipeline.py
│ ├── revenue_attribution.py
│ └── client_report_generator.py
├── conversion-ops/ ← CRO + lead magnet generation
│ ├── SKILL.md
│ ├── cro_audit.py
│ └── survey_lead_magnet.py
├── podcast-ops/ ← Podcast → content factory
│ ├── SKILL.md
│ └── podcast_pipeline.py
├── team-ops/ ← Performance audits + meeting intel
│ ├── SKILL.md
│ ├── team_performance_audit.py
│ └── meeting_action_extractor.py
└── sales-playbook/ ← Value-based pricing framework
├── SKILL.md
├── value_pricing_briefing.py
├── value_pricing_packager.py
├── call_analyzer.py
└── pricing_pattern_library.py
🔒 Privacy & Security
Every skill is built with data privacy in mind:
- PII Sanitizer scans code and data for sensitive information before commits (
security/sanitizer.py)
- Pre-commit hook blocks commits containing detected PII patterns
- Configurable blocklists for company names, person names, and custom patterns
- See
security/README.md for setup
# Scan for sensitive data
python3 security/sanitizer.py --scan --dir . --recursive
# Install the pre-commit hook
cp security/pre-commit-hook.sh .git/hooks/pre-commit && chmod +x .git/hooks/pre-commit
📡 Telemetry (Opt-In)
Anonymous usage telemetry helps us understand which skills people actually use. Fully opt-in, privacy-first:
- Local logging always — see your own usage stats in
~/.ai-marketing-skills/analytics/
- Remote reporting optional — only if you explicitly opt in on first run
- Data collected: skill name, duration, success/fail, version, OS. Nothing else. No code, no file paths, no repo content.
- Version checks — get notified when new skills are available
# View your local usage stats
python3 telemetry/telemetry_report.py
# Check for updates
python3 telemetry/version_check.py
See telemetry/README.md for details.
🤝 Contributing
Found a bug? Have an improvement? PRs welcome. Read CONTRIBUTING.md for guidelines.
- Fork the repo
- Create your feature branch (
git checkout -b feature/better-scoring)
- Run
python3 security/sanitizer.py --scan before committing
- Commit your changes
- Push to the branch
- Open a Pull Request
📄 License
MIT License. Use these however you want.
Star this repo if you find it useful. It helps others discover these tools.
一套面向市场营销和销售团队的开源 Claude Code 技能合集，涵盖增长实验、销售管道自动化、内容运营、外拓获客、SEO、财务运营和视频制作。它们是完整的工作流——包含脚本、评分算法、专家小组和自动化流水线——由 Single Brain 团队构建，并在真正创收的业务管道上经过测试。