openclaw-marketing-skills

# OpenClaw Marketing Skills > Skill by [ara.so](https://ara.so) — Marketing Skills collection. 36 battle-tested marketing skills for OpenClaw agents with real data connectors for Google Ads, Search Console, and Meta Ads. This library turns your AI coding agent into a full-stack marketing brain that reads your codebase, pulls live data, and produces shippable work. ## What This Is A comprehensive skill library that gives OpenClaw agents expertise in: - **CRO** — Conversion rate optimization across signup flows, landing pages, forms, paywalls - **Copywriting** — Homepage copy, email sequences, cold outreach, social content - **SEO** — Technical audits, AI search optimization, programmatic SEO, content strategy - **Paid Ads** — Campaign strategy, creative generation, A/B testing - **Data Connectors** — Live Google Ads, Search Console, and Meta Ads data integration - **Growth & Sales** — Referral programs, launch strategy, pricing, competitor pages ## Installation ### Via ClawHub (Recommended) ```bash # Install all 36 skills clawhub install LeoYeAI/openclaw-marketing-skills # Install specific skills only clawhub install LeoYeAI/openclaw-marketing-skills --skill copywriting page-cro seo-audit # Install data connectors clawhub install LeoYeAI/openclaw-marketing-skills --skill google-ads-connect search-console-connect meta-ads-connect ``` ### Manual Installation ```bash # Clone the repository git clone https://github.com/LeoYeAI/openclaw-marketing-skills.git # Copy skills to OpenClaw directory cp -r openclaw-marketing-skills/skills/* ~/.agents/skills/ # Verify installation ls ~/.agents/skills/ | grep -E "(copywriting|page-cro|seo-audit)" ``` ## Initial Setup ### Step 1: Create Product Context Every skill references your product context automatically. Set this up once: ```bash # In your OpenClaw agent chat Set up my product marketing context ``` This creates `.agents/product-marketing-context.md`: ```markdown # Product Marketing Context ## Product Name [Your product name] ## What It Does [One-line description] ## Target Customer - Role: [e.g., "B2B SaaS founders"] - Pain: [Primary problem you solve] - Stage: [e.g., "0-10 employees, $0-$1M ARR"] ## Positioning [How you're different from alternatives] ## Pricing [Your pricing model and tiers] ## Key Metrics - MRR: [Current monthly recurring revenue] - Conversion Rate: [Visitor to trial/paid] - Churn: [Monthly churn %] ``` ### Step 2: Connect Data Sources (Optional) ```bash # Connect Google Ads Connect my Google Ads account # Connect Search Console Connect Google Search Console # Connect Meta Ads Connect my Meta Ads account ``` This creates credential files in `.agents/`: ``` .agents/ ├── product-marketing-context.md ├── google-ads-credentials.json ├── search-console-credentials.json ├── meta-ads-credentials.json ├── google-ads-data.json # Cached data └── gsc-data.json # Cached data ``` ## Core Skills Reference ### CRO Skills #### page-cro Audits marketing pages for conversion optimization: ```bash # Audit a landing page Audit the conversion rate of /pricing # Audit with specific URL Analyze https://example.com/demo for conversion issues # Output includes: # - Conversion killer diagnosis (scored 1-10) # - Top 3 priority fixes with expected impact # - Before/after copy examples ``` #### signup-flow-cro Optimizes registration and trial flows: ```bash # Audit signup flow Review our signup flow for friction points # Analyze specific steps Check our trial activation flow # Returns: # - Step-by-step friction analysis # - Form field optimization # - Social proof placement recommendations ``` ### Copywriting Skills #### copywriting Generates conversion-focused marketing copy: ```bash # Homepage hero section Write homepage hero copy for [product context] # Feature page Write copy for our [feature name] landing page # Output format: # - Headlines (3 variants) # - Subheadline # - CTA copy # - Benefit bullets # - Social proof section ``` Example output structure: ```markdown ## Headlines 1. [Primary headline - outcome focused] 2. [Alternative - pain point focused] 3. [Alternative - transformation focused] ## Subheadline [Clarifies the promise, adds specificity] ## Primary CTA [Action-oriented, benefit-clear] ## Benefits - [Benefit 1 - specific outcome] - [Benefit 2 - time/money saved] - [Benefit 3 - status/transformation] ``` #### cold-email Generates B2B outreach sequences: ```bash # Create cold email sequence Write a 3-email cold outreach sequence for [target role] # Single email Write a cold email for [specific use case] # Frameworks used: # - PAS (Problem-Agitate-Solution) # - AIDA (Attention-Interest-Desire-Action) # - Problem-first hooks ``` ### SEO Skills #### seo-audit Technical and on-page SEO analysis: ```bash # Full site audit Run SEO audit on example.com # Specific page audit Audit SEO for /blog/post-title # Returns: # - Technical issues (speed, mobile, indexing) # - On-page optimization gaps # - Priority fix list with impact scores ``` #### search-console-connect Pulls live Google Search Console data: ```bash # Diagnose traffic drop Why did organic traffic drop last week? # Find keyword opportunities Show me keywords ranking 4-10 with quick-win potential # Detect cannibalization Check for keyword cannibalization issues # Data pulled: # - Queries, impressions, clicks, CTR, position # - Page-level performance # - Device and country breakdowns ``` Example API usage pattern: ```python # The skill uses this under the hood from google.oauth2 import service_account from googleapiclient.discovery import build credentials = service_account.Credentials.from_service_account_file( '.agents/search-console-credentials.json', scopes=['https://www.googleapis.com/auth/webmasters.readonly'] ) service = build('searchconsole', 'v1', credentials=credentials) # Query last 7 days request = { 'startDate': '2026-05-18', 'endDate': '2026-05-25', 'dimensions': ['page', 'query'], 'rowLimit': 1000 } response = service.searchanalytics().query( siteUrl='sc-domain:example.com', body=request ).execute() ``` ### Paid Ads Skills #### google-ads-connect Connects to Google Ads API for live account data: ```bash # Audit campaigns Audit my Google Ads account and find wasted spend # Keyword analysis Show me keywords with high spend and zero conversions # Campaign scoring Score all active campaigns by performance # Returns: # - 7-dimension campaign scorecard # - Wasted spend identification # - Quality Score diagnostics # - Top 3 priority actions ``` Configuration file structure (`.agents/google-ads-credentials.json`): ```json { "developer_token": "YOUR_DEV_TOKEN", "client_id": "YOUR_CLIENT_ID.apps.googleusercontent.com", "client_secret": "YOUR_CLIENT_SECRET", "refresh_token": "YOUR_REFRESH_TOKEN", "customer_id": "1234567890", "login_customer_id": "1234567890" } ``` #### meta-ads-connect Connects to Meta Marketing API: ```bash # Diagnose performance issues Our Meta ads are getting more expensive. What's happening? # Creative analysis Check for creative fatigue in active campaigns # Audience audit Find audience overlap issues # Analyzes: # - Creative fatigue (frequency, CTR decay) # - Learning Phase status # - Audience overlap # - ROAS by ad set ``` ### Growth Skills #### referral-program Designs referral and affiliate programs: ```bash # Design referral program Create a referral program for [product context] # Outputs: # - Incentive structure (both sides) # - Tracking implementation # - Email templates # - Landing page copy ``` #### marketing-ideas Filters 139 proven SaaS marketing tactics: ```bash # Get ideas by stage Show me marketing ideas for early-stage SaaS # Filter by channel What are the best content marketing tactics? # By goal Give me ideas to reduce churn ``` ## Common Workflows ### Full Landing Page Optimization ```bash # 1. Audit current page Audit the conversion rate of /pricing # 2. Rewrite copy based on audit Write new pricing page copy addressing the issues found # 3. Optimize form Review the signup form on /pricing for friction # 4. Set up A/B test Create an A/B test plan for old vs new pricing page ``` ### SEO + Data Audit Workflow ```bash # 1. Connect data source Connect Google Search Console # 2. Identify issues Run SEO audit on example.com # 3. Find quick wins Show me keywords ranking 4-10 with traffic potential # 4. Create content strategy Build a content strategy targeting those keywords ``` ### Paid Ads Optimization ```bash # 1. Connect account Connect my Google Ads account # 2. Full audit Audit my Google Ads account and find wasted spend # 3. Generate new creative Write 10 ad headline variants for [campaign] # 4. Set up experiment Create an A/B test for the new ad creative ``` ## Environment Variables Never hardcode credentials. Reference environment variables: ```bash # Google Ads export GOOGLE_ADS_DEVELOPER_TOKEN="${GOOGLE_ADS_DEV_TOKEN}" export GOOGLE_ADS_CLIENT_ID="${GOOGLE_ADS_CLIENT_ID}" export GOOGLE_ADS_CLIENT_SECRET="${GOOGLE_ADS_CLIENT_SECRET}" # Meta Ads export META_ADS_ACCESS_TOKEN="${META_ADS_TOKEN}" export META_ADS_AD_ACCOUNT_ID="${META_ADS_ACCOUNT}" # Search Console export GSC_SERVICE_ACCOUNT_JSON="${GSC_CREDENTIALS_PATH}" ``` ## File Structure ``` .agents/ ├── skills/ │ ├── page-cro.md │ ├── copywriting.md │ ├── seo-audit.md │ ├── google-ads-connect.md │ ├── search-console-connect.md │ └── [33 more skills] ├── product-marketing-context.md # Your product info ├── google-ads-credentials.json # API credentials ├── google-ads-data.json # Cached account data └── gsc-data.json # Cached GSC data ``` ## Skill Selection Logic The agent automatically picks the right skill based on your request: | You say | Agent uses | |---------|------------| | "Audit my pricing page" | `page-cro` | | "Write homepage copy" | `copywriting` | | "Why did traffic drop?" | `search-console-connect` | | "Find wasted ad spend" | `google-ads-connect` | | "Optimize signup flow" | `signup-flow-cro` | | "Create email sequence" | `email-sequence` | ## Troubleshooting ### Data Connector Issues **Google Ads connection fails:** ```bash # Verify credentials file exists ls -la .agents/google-ads-credentials.json # Check credential format cat .agents/google-ads-credentials.json | jq . # Test API access manually python -c "from google.ads.googleads.client import GoogleAdsClient; client = GoogleAdsClient.load_from_storage('.agents/google-ads-credentials.json'); print('Connected')" ``` **Search Console "Not authorized":** ```bash # Verify service account has access # 1. Go to Search Console → Settings → Users # 2. Add service account email with "Full" permission # 3. Wait 5 minutes for propagation ``` **Meta Ads token expired:** ```bash # Tokens expire after 60 days # Regenerate at: https://developers.facebook.com/tools/explorer/ # Permissions needed: ads_read, ads_management ``` ### Skill Not Found ```bash # List installed skills clawhub list # Reinstall specific skill clawhub install LeoYeAI/openclaw-marketing-skills --skill page-cro --force # Verify skill file cat ~/.agents/skills/page-cro.md ``` ### Context Not Loading ```bash # Check context file exists cat .agents/product-marketing-context.md # Recreate if missing Set up my product marketing context # Verify skills can read it ls -la .agents/ | grep context ``` ## Advanced Usage ### Custom Skill Combination Chain multiple skills for complex tasks: ```bash # Full marketing stack audit 1. Audit my Google Ads account 2. Review SEO for /pricing 3. Optimize signup flow 4. Write new homepage copy 5. Create A/B test plan for all changes ``` ### Batch Operations ```bash # Audit multiple pages Audit conversion for: /pricing, /features, /demo # Generate variants Write 5 different homepage hero sections # Compare approaches Compare PAS vs AIDA framework for cold email to [persona] ``` ### Data Export ```bash # Export audit results Save the SEO audit results to seo-audit-2026-05.md # Export ad data Export Google Ads performance data to CSV # Share context Create a marketing brief doc from product context ``` ## Related Resources - **MyClaw.ai**: Cloud-hosted OpenClaw with all skills pre-installed - **ClawHub**: Central repository for OpenClaw skills - **OpenClaw Docs**: https://docs.openclaw.ai ## Contributing To add or improve skills: ```bash # Fork and clone git clone https://github.com/YOUR_USERNAME/openclaw-marketing-skills.git # Create new skill cp skills/template.md skills/your-skill-name.md # Test locally cp skills/your-skill-name.md ~/.agents/skills/ # Test with OpenClaw agent # Submit PR git checkout -b add-your-skill git commit -am "Add [skill-name] skill" git push origin add-your-skill ``` ## License MIT License — See LICENSE file in repository. --- **Powered by [MyClaw.ai](https://myclaw.ai)** — Run all 36 skills without managing a server.