CardPilot Remote MCP Server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
CardPilot Remote MCP Server
This is a remote MCP (Model Context Protocol) server for CardPilot. It exposes tools to fetch and analyze credit card data.
Server URL
Public URL: https://mcp.cardpilot.ca/sse
Note: This server uses a custom domain to ensure compatibility with OpenAI Agent Builder.
Related MCP server: KoKo Finance MCP Server
Available Tools
1. get-cards
Fetches a list of credit cards with detailed metadata, suitable for ranking and comparison.
Input parameters:
Parameter | Type | Description |
|
| Sort criteria: |
|
| Sort direction: |
|
| Comma-separated list of card IDs (e.g., |
|
| Filter by bank name (e.g., "TD", "RBC") |
|
| Filter by category (e.g., "travel", "cash back") |
|
| Set to |
|
| Maximum number of cards to return (default: 5) |
|
| Target persona for tailored ranking: |
Output structure:
{
"cards": [
{
"cardId": "string",
"name": "string",
"bank": "string",
"annualFee": number,
"score": number,
"details": { "rewards": number, "perks": number, "fees": number, ... }
}
],
"meta": { "total": number, "sort": "string", "direction": "string" }
}2. get-guides
Fetches a list of educational guide metadata. This tool is optimized for chatbots to provide links to full articles.
Input parameters: None
Output structure:
{
"guides": [
{
"slug": "string",
"title": "string",
"intro": "string",
"icon": "string (emoji)",
"url": "string"
}
]
}Agent System Prompt
If you are using this MCP server with an LLM Agent (like OpenAI Custom GPT), add the following to your System Instructions:
You are an expert credit card advisor powered by CardPilot data.
1. **Card Recommendations**:
- ALWAYS use the `get-cards` tool to fetch real-time data before making recommendations.
- Set `limit=5` by default to ensure concise recommendations.
- If the user specifies or asks for a specific persona (e.g., student, newcomer), set the `persona` parameter (e.g., `persona="student"`).
- Use the `sort` parameter to align with user priorities (e.g., `sort="annual_fee"` for cheap cards).
- Use filters like `bank="TD"` or `category="travel"` to narrow down results.
- For "no fee" requests, explicitly set `noFee=true`.
- When presenting cards, list the Name, Annual Fee, Welcome Bonus, and a brief "Why it fits" explanation.
- Link the card name to the `applyUrl` or CardPilot detail page if available.
2. **Educational Content**:
- If a user asks general questions (e.g., "Cash back vs Points"), use `get-guides` to see if there is a relevant article.
- Provide the answer based on the guide's `intro` and encourage the user to read the full guide by providing the `url`.
3. **General Rules**:
- Do not make up card details. If data is missing, state that.
- Be concise and helpful.How to Use
OpenAI Agent Builder
Create a new Agent.
Under Actions, click Add Action.
Select "Add from URL".
Enter:
https://mcp.cardpilot.ca/sseIt should load the
get-cardstool immediately.
Cloudflare AI Playground
Enter the server URL:
https://mcp.cardpilot.ca/sseClick Connect.
Claude Desktop
Add the following to your claude_desktop_config.json:
{
"mcpServers": {
"cardpilot": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.cardpilot.ca/sse"
]
}
}
}Troubleshooting
OpenAI 424 Error ("Unable to load tools")
If you see this error, ensure you are using the Custom Domain URL (mcp.cardpilot.ca).
OpenAI blocks generic .workers.dev domains for MCP servers.
Local Development
To test locally:
Install dependencies:
npm installStart local server:
npm start(Runs on port 8787)
Expose via Ngrok (Required for OpenAI testing):
ngrok http 8787Use the Ngrok URL (e.g., https://xxxx.ngrok-free.app/sse) in Agent Builder.
Deployment
Deploy to Cloudflare Workers:
npm run deployThe server will be live at https://mcp.cardpilot.ca/sse.
This server cannot be deployed
Maintenance
Related MCP Connectors
AI credit card advisor - search cards, compare portfolios, and optimize rewards
Search, compare and estimate rewards across US credit cards. Verified fees, APRs and bonuses.
Which of your credit cards to use at any store, from verified bank data. No bank login.
US credit card data: annual fees, credit card credits, earn rates, offers and cashback portal rates.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to search, compare, and get personalized recommendations for financial products like savings accounts, credit cards, loans, budgeting apps, and earning platforms using curated data from WalletGrower.1MIT
- AlicenseNot gradedqualityBmaintenanceA hosted MCP server for AI-powered credit card advice, enabling search, comparison, portfolio analysis, and personalized recommendations for US credit cards through natural language.1MIT
- AlicenseAqualityCmaintenanceEnables agents to recommend the best credit card for a given store or category, look up merchant coding, search cards, and inspect rotating bonus calendars using verified reward data.51,195 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables users to search and compare credit card recommendations, view card details and popular top-5 cards, and retrieve financial knowledge articles through MCP tools.-