Skip to main content
In addition to tools, the Orchata MCP server provides resources for documentation access and prompts for guided RAG workflows.

​
Resources

MCP resources provide read-only access to documentation and dynamic data. AI assistants can read these to understand how to use Orchata.

​
Documentation Resources

URIDescription
orchata://docs/introductionPlatform introduction and core concepts
orchata://docs/quickstartGetting started guide for AI agents
orchata://docs/api-referenceAPI and MCP tool reference
Example usage: 
Read the resource at orchata://docs/quickstart
This returns a Markdown document explaining how to use Orchata step-by-step.

​
Dynamic Space Resources

Access detailed information about any space:
URI PatternDescription
orchata://spaces/{spaceId}Detailed space information
Example: 
Read the resource at orchata://spaces/spc_abc123
Returns JSON with the space’s full details including document count, description, and metadata.
When an MCP client requests the list of resources, all available spaces are automatically included in the response.

​
Prompts

MCP prompts provide guided workflows for common RAG tasks. They help AI assistants understand best practices and follow optimal patterns.

​
rag-workflow

Step-by-step guide for implementing a RAG workflow with Orchata. Parameters:
​
useCase
string
Type of RAG application: chatbot, search, qa, or agent
Example: 
Use the rag-workflow prompt with useCase="agent"
Returns guidance on:
  1. Organizing knowledge into spaces
  2. Choosing query strategies
  3. Formatting context for LLMs
  4. Generating grounded responses
Sample output for agent use case: 
# Orchata RAG Workflow Guide

## Basic RAG Flow

1. **Organize Knowledge**: Create focused spaces for different topics
2. **Query Strategy**: Use smart_query when unsure, query_spaces when you know the space
3. **Context Formatting**: Include retrieved content in your LLM prompt
4. **Response Generation**: Generate responses grounded in retrieved content

## AI Agent Tips
- Always start with smart_query to find relevant spaces
- Search multiple spaces for comprehensive answers

​
space-discovery

Help discover which spaces contain the information you need. Parameters:
​
topic
string
The topic or question you’re trying to find information about
Example: 
Use the space-discovery prompt with topic="authentication best practices"
Returns guidance on:
  1. How to use smart_query for the topic
  2. How to search the recommended spaces
  3. Strategies for refining results
Sample output: 
# Space Discovery Guide

Looking for: **authentication best practices**

1. Use smart_query("authentication best practices")
2. Then query_spaces with returned IDs

​
Using Resources and Prompts

​
When to Use Resources

  • Learning: AI needs to understand how Orchata works
  • Reference: AI needs to look up API details or concepts
  • Context: AI needs information about a specific space

​
When to Use Prompts

  • Guidance: AI needs help with the right approach
  • Best Practices: AI needs to follow recommended patterns
  • Discovery: AI needs to find relevant information

​
Example Workflow

Here’s how an AI assistant might use resources and prompts together:
1

Read Documentation

AI reads orchata://docs/introduction to understand core concepts.
2

Use Discovery Prompt

AI uses space-discovery prompt with the user’s topic.
3

Execute Smart Query

AI calls smart_query tool to find relevant spaces.
4

Search Spaces

AI calls query_spaces with discovered space IDs.
5

Generate Response

AI synthesizes an answer from the retrieved content.

​
Resource and Prompt Reference

​
All Resources

URITypeDescription
orchata://docs/introductionStaticPlatform overview
orchata://docs/quickstartStaticGetting started guide
orchata://docs/api-referenceStaticAPI reference
orchata://spaces/{id}DynamicSpace details

​
All Prompts

NameArgumentsDescription
rag-workflowuseCase?RAG implementation guide
space-discoverytopic?Space discovery guide

​
Next Steps

MCP Tools

Explore all 12 MCP tools.

Setup Guides

Configure your AI tool for Orchata.