TarotAPI MCP. Access structured card data and multi-card readings.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
TarotAPI connects your AI client directly to a complete, structured 78-card tarot deck. It lets you pull random cards for daily guidance or run complex multi-card spreads—like the Celtic Cross—and get full interpretations on demand.
You can also search the whole deck using specific keywords or meanings.
What your AI agents can do
Get api info
Retrieves basic meta-information about the TarotAPI server setup.
Get card
Fetches detailed data for a specific card using its short name, including upright and reversed meanings and keywords.
Get multiple random cards
Pulls several random cards simultaneously; this is useful for structured spreads like the Celtic Cross layout.
Retrieves a card's full data—keywords, description, upright/reversed meaning, and image—given its short name (e.g., 'ar01').
Pulls multiple random cards for structured readings like the Celtic Cross or simple three-card sequences.
Gets one random card suitable for quick, daily reflection without needing specific inputs.
Finds all cards in the deck that match a given text query, whether it's an upright meaning or general keyword.
Returns a full inventory of all 78 cards, including their names and unique short identifiers for later use.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
TarotAPI MCP Server: 6 Tools for Tarot Readings
These tools let your AI agent interact with the full tarot deck. You can pull random cards, search by keywords, or retrieve specific card details in a structured format.
019d8487get api info
Retrieves basic meta-information about the TarotAPI server setup.
019d8487get card
Fetches detailed data for a specific card using its short name, including upright and reversed meanings and keywords.
019d8487get multiple random cards
Pulls several random cards simultaneously; this is useful for structured spreads like the Celtic Cross layout.
019d8487get random card
Generates a single, completely random card suitable for quick daily guidance or reflection.
019d8487list all cards
Returns the full catalog of all 78 cards, listing their names, numbers, and available identifiers.
019d8487search cards
Searches the entire deck based on general text queries or specific meanings (upright/reversed).
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with TarotAPI, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
You gotta hook your AI client up with TarotAPI to get access to a complete, structured 78-card tarot deck. Forget general knowledge—this server gives you verified data for every single major and minor arcana card. When you run these tools, you're not just getting vague vibes; you’re pulling specific interpretations and keywords directly from the API structure.
To start up, you can use get_api_info to pull basic meta-information on how the server is set up. If you need a full inventory of what’s in the deck before running anything else, run list_all_cards. That tool returns the total catalog of all 78 cards, listing their official names, assigned numbers, and unique short identifiers that you'll need later.
When it comes to reading specific cards, your options are pretty deep. If you know exactly which card you want—say, The Tower or The Lovers—you use get_card. Just feed it the card’s short name (like 'ar01'), and it spits out all the detailed info: keywords, a full description, its meaning when upright, and what it means if it's reversed.
It gives you the whole package.
If you need to find cards based on something other than their name—maybe you’re looking for a card related to 'change' or 'communication'—you use search_cards. This tool searches the entire deck using general text queries, letting you pinpoint every card whose meaning or keyword matches what you type in. You can search by upright meanings or reversed interpretations.
For quick reads, you don’t gotta do much work. If you just want a single random card for daily reflection, run get_random_card. It immediately generates one completely arbitrary card, perfect for quick guidance without needing any inputs from your side. When you need more than one card—like running a full Celtic Cross spread or even just looking at past/present/future sequences—you call get_multiple_random_cards.
This tool pulls several random cards all at once, letting your agent handle structured layouts that require multiple specific placements.
These tools mean you control the entire reading process. You use list_all_cards to build your knowledge base; then you pull a single card with get_card; or if you're doing a spread, you hit it with get_multiple_random_cards. If all that fails, you can always narrow things down by querying the data using search_cards, making sure every interpretation you pass to your user is backed up by structured data.
You don’t have to guess; you just call the right tool.
How TarotAPI MCP Works
- 1 First, your AI agent uses
list_all_cardsto confirm the available card identifiers (e.g., 'swac', 'ar01'). - 2 Next, you prompt the agent with a request—for example, "Give me a three-card spread.". The agent then calls
get_multiple_random_cards. - 3 You receive structured JSON output containing the card names, their associated meanings (upright/reversed), and keywords for immediate use.
The bottom line is that you get clean, structured tarot data without having to manually cross-reference 78 cards across multiple websites or databases.
Who Is TarotAPI MCP For?
This tool is for content creators, narrative designers, and technical writers who need consistent, symbolic data. It targets the pain of manual research: the person writing a weekly 'wisdom' column who hates spending hours cross-referencing card meanings across different sources.
Uses get_random_card for quick daily blog posts or uses search_cards to find specific themes (like 'courage') and pull relevant supporting cards.
Integrates the API into character sheet generation, using get_multiple_random_cards to create structured plot devices or destiny readings for new campaigns.
Builds interactive learning modules where users can click on a card and have the agent fetch its full details using get_card, ensuring data integrity across all interpretations.
What Changes When You Connect
- Structured Interpretation: Instead of vague text, the API returns dedicated fields for Upright Meaning, Reversed Meaning, Keywords, and general descriptions. This consistency is critical when building reliable content loops.
- Versatile Readings: The
get_multiple_random_cardstool lets you run complex spreads (like Past/Present/Future) in one call. You get a structured reading immediately, not just a list of cards. - Precision Search: Use
search_cardsto find every card that mentions a specific concept—say, 'challenge' or 'growth.' This is better than searching a general knowledge base because you are confined to the 78-card canon. - Efficiency for Daily Content: The
get_random_cardtool eliminates setup time. It provides fresh, guided content instantly, perfect for daily email digests or blog prompts. - Full Deck Mapping: Running
list_all_cardsgives you all the necessary identifiers and metadata to build a complete front-end browsing experience without guessing card codes.
Real-World Use Cases
Creating Daily Devotional Content
A wellness app needs daily content. Instead of writing a new reading every day, the agent calls get_random_card to pull today's card. It then passes that single card to the AI client, which formats the keywords and meaning into a ready-to-post article.
Designing RPG Quests
A game designer needs a plot device. They ask the agent for a 'three-card spread' using get_multiple_random_cards. The resulting Past/Present/Future cards become three mandatory elements of the quest narrative, ensuring symbolic consistency.
Building an Educational Quiz Tool
A student needs to learn card meanings. They use list_all_cards first, then prompt the agent to 'search for all cards related to finance.' The tool runs search_cards, giving a clean list of relevant concepts and associated cards.
Debugging Card Identifiers
A developer is building an inventory viewer. They start by running list_all_cards just to ensure they have the correct short names (like 'swac' vs. 'Ace of Cups'), preventing runtime errors when calling get_card.
The Tradeoffs
Treating it like a general search engine
User prompts: 'Tell me about love and rebirth.' The AI might give generic spiritual platitudes that don't reference any specific card or structure.
→
To get structured data, use search_cards and specify the intent. For instance: 'Search for cards matching keywords: 'rebirth' OR 'love'. This forces the system to pull only from the 78-card canon.'
Calling multiple tools unnecessarily
User calls get_random_card and then immediately calls search_cards for general keywords. This is redundant if the goal was just daily reflection.
→
If you need a simple, quick reading, stick to get_random_card. Only use search_cards when you have a specific keyword or meaning you need to cross-reference across multiple cards.
Assuming full card data is available
The user asks for the 'ruling history' of The Tower. This metadata isn't in the core dataset, and calling get_card won't provide it.
→
Only expect detailed interpretations (upright/reversed meaning) from get_card. If you need external data like rulings or market prices, use a different API service.
When It Fits, When It Doesn't
Use this MCP Server if your core requirement is structured symbolic interpretation: You need specific fields for Upright Meaning, Reversed Meaning, and keywords that are tied to the 78-card canon. If you're building an educational tool or a narrative engine that requires consistent data output, this is what you want.
Don't use it if your goal is simply 'chat.' If you just want to chat with an AI about tarot history and theories—without needing the specific meanings of The Lovers (VI) versus The Hierophant (V)—then a standard text model will suffice. This tool mandates structured interaction via its six tools; it's not for open-ended conversation.
If you need general card data, start with list_all_cards. If you need specific details, use get_card(short_name). For complex readings, always default to get_multiple_random_cards.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by TarotAPI. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 6 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding a consistent interpretation for cards shouldn't require cross-referencing six different websites.
Currently, if you need to know what 'abundance' means across three different spreads—say, past, present, and future—you have to open multiple tabs. You pull up a basic meaning for the card, then search for its keywords in a second place, and finally check a third site just for the reversed definition. It’s manual, time-consuming data aggregation.
With TarotAPI, your agent handles this logic. Whether you use `get_multiple_random_cards` or run a single `get_card`, all three required meanings—upright meaning, keywords, and fortune telling interpretation—are returned in one clean JSON payload. You get the full context immediately.
TarotAPI MCP Server: Get structured card guidance instantly.
Before this server, generating a 'Past/Present/Future' reading was a multi-step process. You had to manually draw three cards, look up each one individually for its meaning, and then write the connection between them. The output was always disjointed.
Now, running `get_multiple_random_cards` provides all three interpretations—the Past, Present, and Future—structured under specific headers in a single API call. It eliminates the guesswork and gives you clean, ready-to-use narrative blocks.
Common Questions About TarotAPI MCP
How do I get card details for a specific card using TarotAPI? +
Use the get_card tool. You must provide the short name (like 'ar01' or 'swac'). This returns all structured data, including both upright and reversed meanings.
Can I run a three-card spread using get_multiple_random_cards? +
Yes. get_multiple_random_cards is designed for this. It pulls multiple random cards simultaneously, making it perfect for structured layouts like Past/Present/Future.
What if I want to find all cards related to 'courage'? +
Use search_cards. You can query the tool with keywords or specific meanings. This will return every card in the deck that matches your search criteria.
Is there a way to see what cards are available for me to use? +
Run list_all_cards. This returns a complete inventory of all 78 tarot cards, giving you their names and the unique short identifiers needed for other tools.
Does TarotAPI support reversed card meanings? +
Yes. The get_card tool explicitly includes fields for both upright and reversed interpretations for every major arcana card.
Using the get_random_card tool, how can I pull a single card for quick daily guidance? +
The tool instantly returns one random tarot card. The response provides the full meaning, keywords, and interpretations right away. This is perfect when you need immediate reflection or general advice without specifying a particular card.
When I use the get_api_info tool, what technical details about TarotAPI should I expect? +
It provides essential metadata for the API. This information helps your AI client understand the service's structure and available endpoints before making a functional call.
If I try to run the get_card tool with an invalid short name, what error do I receive? +
The system returns a clear error message. This response specifies that the card identifier is incorrect and guides you toward using valid formats like 'ar01' or 'swac'.
Do I need an API key or account? +
No! TarotAPI is completely free and open with no authentication required.
How many cards are in the deck? +
The deck contains all 78 traditional tarot cards: 22 Major Arcana and 56 Minor Arcana (Wands, Cups, Swords, Pentacles).
Can I do a three-card past/present/future reading? +
Yes! Use get_multiple_random_cards with count=3 to draw three cards for a past/present/future spread.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Time-Series Seasonality Engine
Compute exact Autocorrelation (ACF) to find seasonality lags in time-series data without hallucination.
OpenAI
Use GPT-4o, DALL-E 3, embeddings, fine-tuning, and moderation as tools inside your AI agent workflows.
FlowiseAI
Build LLM orchestration flows visually with a drag-and-drop interface for creating AI chatbots, agents, and RAG pipelines.
You might also like
Vapi
Command Voice AI assistants directly from your chat. Make outbound phone calls, update personas, and retrieve full transcripts via Vapi.
Birdeye
Manage reputation and customer experience via Birdeye — list reviews, manage contacts, and trigger surveys directly from any AI agent.
Octoparse
Connect your AI agent to Octoparse to trigger cloud web scraping tasks, monitor crawler statuses, and retrieve scraped data directly into chat.