Add ourdigital-seo-audit skill with: - Full site audit orchestrator (full_audit.py) - Google Search Console and PageSpeed API clients - Schema.org JSON-LD validation and generation - XML sitemap and robots.txt validation - Notion database integration for findings export - Core Web Vitals measurement and analysis - 7 schema templates (article, faq, product, etc.) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
5.3 KiB
5.3 KiB
Claude Desktop SEO Audit - Setup Guide
Prerequisites
1. Claude Desktop Application
- Download from https://claude.ai/download
- Sign in with your Anthropic account
- Pro subscription recommended for extended usage
2. Required MCP Servers
Configure these MCP servers in Claude Desktop settings:
Firecrawl (Web Scraping)
{
"mcpServers": {
"firecrawl": {
"command": "npx",
"args": ["-y", "firecrawl-mcp"],
"env": {
"FIRECRAWL_API_KEY": "your-firecrawl-api-key"
}
}
}
}
Get API key from: https://firecrawl.dev
Notion (Database Storage)
{
"mcpServers": {
"notion": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-notion"],
"env": {
"NOTION_API_KEY": "your-notion-api-key"
}
}
}
}
Get API key from: https://www.notion.so/my-integrations
Perplexity (Research)
{
"mcpServers": {
"perplexity": {
"command": "npx",
"args": ["-y", "perplexity-mcp"],
"env": {
"PERPLEXITY_API_KEY": "your-perplexity-api-key"
}
}
}
}
Get API key from: https://www.perplexity.ai/settings/api
3. Notion Database Setup
Option A: Use Existing Database
The OurDigital SEO Audit Log database is already configured:
- Database ID:
2c8581e5-8a1e-8035-880b-e38cefc2f3ef
Ensure your Notion integration has access to this database.
Option B: Create New Database
Create a database with these properties:
| Property | Type | Options |
|---|---|---|
| Issue | Title | - |
| Site | URL | - |
| Category | Select | Technical SEO, On-page SEO, Content, Local SEO, Performance, Schema/Structured Data, Sitemap, Robots.txt |
| Priority | Select | Critical, High, Medium, Low |
| Status | Status | Not started, In progress, Done |
| URL | URL | - |
| Found Date | Date | - |
| Audit ID | Rich Text | - |
Configuration File Location
macOS
~/Library/Application Support/Claude/claude_desktop_config.json
Windows
%APPDATA%\Claude\claude_desktop_config.json
Linux
~/.config/Claude/claude_desktop_config.json
Complete Configuration Example
{
"mcpServers": {
"firecrawl": {
"command": "npx",
"args": ["-y", "firecrawl-mcp"],
"env": {
"FIRECRAWL_API_KEY": "fc-your-key-here"
}
},
"notion": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-notion"],
"env": {
"NOTION_API_KEY": "ntn_your-key-here"
}
},
"perplexity": {
"command": "npx",
"args": ["-y", "perplexity-mcp"],
"env": {
"PERPLEXITY_API_KEY": "pplx-your-key-here"
}
},
"fetch": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-fetch"]
}
}
}
Adding Project Knowledge
Step 1: Create a Project
- Open Claude Desktop
- Click on the project selector (top left)
- Click "New Project"
- Name it "SEO Audit"
Step 2: Add Knowledge Files
- In your project, click the paperclip icon or "Add content"
- Select "Add files"
- Add these files from
~/.claude/desktop-projects/seo-audit/:SEO_AUDIT_KNOWLEDGE.md(main knowledge file)QUICK_REFERENCE.md(quick commands)
Step 3: Verify Setup
Start a new conversation and ask:
What SEO audit capabilities do you have?
Claude should describe the available audit features.
Testing the Setup
Test 1: Firecrawl
Scrape https://example.com and show me the page structure
Test 2: Notion
Search for "SEO Audit" in Notion
Test 3: Perplexity
Research current SEO best practices for 2024
Test 4: Full Audit
Perform a quick SEO audit for https://blog.ourdigital.org
Troubleshooting
MCP Server Not Connecting
- Restart Claude Desktop
- Check config file JSON syntax
- Verify API keys are correct
- Check Node.js is installed (
node --version)
Notion Permission Error
- Go to Notion integration settings
- Add the database to integration access
- Ensure integration has read/write permissions
Firecrawl Rate Limit
- Wait a few minutes between requests
- Consider upgrading Firecrawl plan
- Use
firecrawl_mapfor discovery, then targeted scrapes
Knowledge Files Not Loading
- Ensure files are in supported formats (.md, .txt)
- Keep file sizes under 10MB
- Restart the project conversation
Usage Tips
- Start with Quick Reference - Use the commands from QUICK_REFERENCE.md
- One site per conversation - Keep context focused
- Export regularly - Save findings to Notion frequently
- Check existing findings - Query Notion before creating duplicates
- Prioritize Critical issues - Fix showstoppers first
Differences from Claude Code Version
| Feature | Claude Code | Claude Desktop |
|---|---|---|
| Python Scripts | Direct execution | Not available |
| Google APIs | Service account auth | Manual or via MCP |
| File System | Full access | Limited to uploads |
| Automation | Bash commands | MCP tools only |
| Scheduling | Possible via cron | Manual only |
Support
For issues with:
- Claude Desktop: https://support.anthropic.com
- Firecrawl: https://docs.firecrawl.dev
- Notion API: https://developers.notion.com
- MCP Protocol: https://modelcontextprotocol.io