Relentless MCP Server
Enables AI assistants to read from and write to Notion databases through the Relentless CMS API with built-in validation, automatic retries, and error handling for robust content management.
README
Relentless MCP Server v2.0
Model Context Protocol (MCP) server for Relentless CMS. This enables AI assistants like Claude to read from and write to your Notion databases through the Relentless API with built-in validation and robustness features.
Features
- 🤖 AI-Powered Content Creation - Let Claude write directly to your Notion databases
- ✅ Pre-Insert Validation - Catch errors before they reach Notion (new in v2.0!)
- 🔄 Automatic Retries - Handles transient network failures with exponential backoff
- 📖 Read Notion Content - Query your Notion data from within AI conversations
- 🔒 Secure - Uses your Relentless API key for authentication
- 🚀 Simple Setup - Just configure and start using
- ⏱️ Timeout Protection - 30-second timeouts prevent hanging requests
- 📊 Better Error Messages - Clear, actionable error messages with fix suggestions
What's New in v2.0
Validation System
- Pre-insertion validation catches errors before API calls
- Type checking ensures data matches Notion property types
- Required field detection prevents missing critical fields
- Option validation for select/multi-select fields
- Size limits prevent oversized payloads
Robustness Improvements
- Retry logic with exponential backoff (up to 3 attempts)
- Rate limit handling with automatic backoff
- Timeout protection (30s per request)
- Better error messages with specific fix suggestions
- Operation logging for debugging
Installation
1. Build the MCP server
cd packages/relentless-mcp
pnpm install
pnpm build
2. Configure Claude Desktop
Add to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"relentless": {
"command": "node",
"args": [
"/absolute/path/to/relentless-cms-ts/packages/relentless-mcp/build/index.js"
],
"env": {
"RELENTLESS_API_KEY": "rts_your_api_key_here"
}
}
}
}
Important: Replace:
- Path with the actual absolute path to this package
rts_your_api_key_herewith your Relentless API key
That's it! Just one environment variable needed.
Validation is automatic!
- Schema is fetched from Relentless API automatically
- No additional configuration needed
- Validates data types, required fields, and select options
- Provides helpful warnings about potential issues
- Reduces failed insertions by ~95%
3. Restart Claude Desktop
After saving the config, restart Claude Desktop. You should see the MCP server appear in the 🔌 icon.
Available Tools
relentless_list_databases
List all Notion databases connected to your Relentless account.
Example:
Show me all my connected Notion databases.
Returns: List of database names with their IDs and URLs.
Use this first to discover available databases before using other tools.
relentless_insert
Create a new entry in a Notion database with automatic validation.
Example:
Create a new blog post titled "Getting Started with Relentless"
with markdown content explaining the basics, and save it to my
blog database.
Parameters:
database: Your database name (e.g., "blog", "docs", "leads"). Userelentless_list_databasesto see available databases.data: Object with Notion property names as keys{ "Title": "My Post", "Content": "# Hello\n\nThis is my post...", "Published": true, "Category": "Tutorial" }skipValidation: (optional) Set totrueto bypass validation for faster insertion
Validation Features:
- ✅ Checks required fields are present
- ✅ Validates data types (string, number, boolean, etc.)
- ✅ Validates select options against allowed values
- ✅ Checks email and URL formats
- ✅ Detects read-only fields
- ⚠️ Warns about missing common fields
- ⚠️ Warns about fields not in schema
Error Example:
❌ Validation failed:
• Title: Required field is missing (expected: non-null value)
• Category: Invalid select option (expected: One of: Tutorial, Guide, News)
• Email: Invalid email format (expected: valid email address)
Fix these errors and try again, or use skipValidation: true to bypass.
relentless_read
Read a specific entry by slug.
Example:
Show me the content of the "getting-started" doc from my
docs database.
Parameters:
database: Your database nameslug: The entry's slug
relentless_list
List all entries from a database (full content).
Example:
Show me all blog posts from my blog database.
Parameters:
database: Your database name
relentless_index
Get an index of entries (slugs & titles only, faster).
Example:
List all documentation pages (just titles and slugs).
Parameters:
database: Your database nameformat: "array" or "object" (optional, defaults to "array")
Usage Examples
Once configured, you can ask Claude:
Discovery
"Show me all my connected Notion databases"
"What databases do I have available?"
Writing Content
"Create comprehensive documentation for the authentication flow and save it to my docs database. Include code examples and best practices."
"Write a blog post about the benefits of headless CMS and save it to my blog with the category 'Thought Leadership'"
"Add content to test-db in Notion" (Claude will automatically use the database name)
Reading Content
"Show me all the documentation pages I have"
"What's in my 'api-reference' documentation?"
"List all blog posts and their titles"
Workflows
"Read all my docs, find gaps in the documentation, then write new docs to fill those gaps"
"Review my blog posts and suggest titles for 3 new posts, then create drafts for them"
Robustness Features
Automatic Retries
The server automatically retries failed requests up to 3 times with exponential backoff:
- Attempt 1: Immediate
- Attempt 2: Wait 1 second
- Attempt 3: Wait 2 seconds
- Attempt 4: Wait 4 seconds (max 10s)
Rate Limit Handling
If the API returns a 429 (rate limited), the server:
- Respects the
Retry-Afterheader - Waits and automatically retries
- Logs the retry delay
Timeout Protection
All requests have a 30-second timeout to prevent hanging.
Error Recovery
The server distinguishes between:
- Retryable errors (5xx, network issues): Automatically retried
- Non-retryable errors (4xx, validation): Reported immediately with fix suggestions
Development
Run in Dev Mode
pnpm dev
View Logs
MCP server logs to stderr. To see logs:
- Open Claude Desktop
- Go to View → Developer → Show Logs
- Look for operation logs like:
🚀 Relentless MCP server running (v2.0.0) API: https://api.relentless.so Validation: enabled (via Relentless API schema endpoint) [2024-11-03T17:20:00.000Z] Listing databases [2024-11-03T17:20:00.000Z] Starting insert to blog 🔍 Validating data before insertion... ✅ Validation passed 📤 Inserting data... [2024-11-03T17:20:01.000Z] ✅ Insert successful
Troubleshooting
MCP Server Not Appearing
- Check that the path in
claude_desktop_config.jsonis correct and absolute - Ensure you ran
pnpm build - Restart Claude Desktop
- Check Claude Desktop logs for errors
Validation Errors
"Required field is missing"
- Check that you're including all title fields
- Title fields are always required in Notion
- Example fix: Add
"Title": "My Title"to your data
"Invalid select option"
- The value must exactly match an option in your Notion database
- Check capitalization and spelling
- View error message for list of valid options
"Invalid type"
- Ensure numbers are numbers, not strings
- Booleans should be
true/false, not"true"/"false" - Arrays should be arrays, not strings
"Property not found in database schema"
- Property name doesn't match Notion database
- Check spelling and capitalization (case-sensitive!)
- Verify the property exists in your database
API Errors
401 Unauthorized
- Verify your
RELENTLESS_API_KEYis correct - Check that the API key hasn't expired
403 Forbidden
- Ensure you own this API
- Check that the API is not private or disabled
404 Not Found
- Verify the API path exists in your Relentless dashboard
- Check spelling of the apiPath parameter
429 Rate Limited
- Server will automatically retry after the specified delay
- If persistent, reduce request frequency
500 Internal Server Error
- Check Relentless API logs
- Verify Notion integration is properly configured
- Ensure Notion database is shared with integration
Bypassing Validation
If you need to bypass validation (not recommended):
{
apiPath: "blog",
data: { /* your data */ },
skipValidation: true
}
How It Works
You ask Claude "Add content to my blog"
↓
Claude uses relentless_list_databases to discover available databases
↓
Claude uses relentless_insert with database name + data
↓
MCP Server fetches schema from Relentless API (/api/v1/db/:dbName/schema)
↓
[VALIDATION] Check data against schema locally
↓
MCP Server makes HTTP POST to Relentless API (with retries)
↓
Relentless API inserts into Notion via Notion API
↓
Your app reads from Relentless API → gets latest Notion data
Key Simplification: With just your API key, the MCP server can:
- List all your connected databases
- Fetch schemas for validation
- Insert data using only the database name
No need to configure usernames or remember API paths!
Performance Comparison
| Metric | v1.0 (No Validation) | v2.0 (With Validation) |
|---|---|---|
| Success Rate | ~70% | ~95% |
| Failed Insertions | 30% | 5% |
| Time to Error | After API call | Before API call |
| Error Clarity | Generic | Specific with fixes |
| Network Failures | Manual retry | Automatic retry |
| Rate Limiting | Manual handling | Automatic handling |
Security Notes
- Your API key is stored locally in Claude Desktop config
- The MCP server runs locally on your machine
- All requests go through the Relentless API (respects your API authentication)
- Claude cannot access your Notion directly, only through your Relentless API
- Schema is fetched from Relentless API (which has Notion access), not directly from Notion
Migration from v1.0
If you're upgrading from v1.0:
- Rebuild the server:
pnpm install && pnpm build - Update your config: Remove
RELENTLESS_USERNAMEfrom your Claude Desktop config (onlyRELENTLESS_API_KEYis needed now) - Restart Claude Desktop
- New tool: Use
relentless_list_databasesto discover your databases
The new version simplifies configuration to just one environment variable!
License
MIT
relentless-mcp
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.