MCP Tools Overview

Qoris provides 7 powerful MCP (Model Context Protocol) tools that enable AI agents to interact with workspace memory and knowledge. These tools are available as connections that can be configured in any MCP-compatible AI application. MCP is the access layer. Knox is the governance layer. When agents only retrieve context, MCP may be enough. When agents execute commands, call external tools, schedule work, or mutate business systems, Knox should govern the action path.

Memory Tools

4 tools for managing persistent workspace memories that persist across conversations

Knowledge Tools

3 tools for searching, listing, and reading your project’s knowledge base

Knox Governance

Safety checks, hooks, policy enforcement, and audit around agent actions

What is MCP?

Model Context Protocol (MCP) is an open standard that enables AI applications to securely access external data sources and tools. Qoris implements MCP to provide seamless integration with AI assistants, chatbots, and automated workflows. In Qoris, MCP connections are intended to work with the broader governed-agent stack:

MCP provides the interface to memory, knowledge, and external tools
Knox checks risky actions before execution
approvals handle business judgment for sensitive operations
audit records the path agents took

Authentication

All MCP tools require authentication via API keys. Each API key is scoped to a specific project and user, ensuring secure access to workspace data.

Generate API keys in your Qoris project settings. Each key provides access to all tools for the associated project.

Available Tools

Memory Tools

Tools for managing persistent workspace memories that persist across conversations:

save_memory

Save important information to workspace memory. Perfect for storing user preferences, decisions, and context.

get_memories

Retrieve all workspace memories with pagination support. Load context at the start of conversations.

update_memory

Update existing memories. Correct information or add additional context as needed.

delete_memory

Delete memories from the workspace. Remove outdated or incorrect information.

Knowledge Tools

Tools for searching, listing, and reading the project’s knowledge base:

search_knowledge

Search across all project knowledge with AI-powered semantic search and synthesis.

list_knowledge_documents

List all documents and webpages in the knowledge base.

get_document_full_content

Retrieve the full raw text of a document by name.

Data Persistence

Memories

Storage: PostgreSQL database with ChromaDB indexing for semantic search
Scoping: All operations are scoped to the project associated with the API key
Ownership: Memories are tied to the API key that created them
Persistence: Memories persist across all conversations in the workspace

Knowledge

Sources: Documents, webpages, and knowledge items uploaded to the project
Search: Semantic search powered by advanced AI models
Synthesis: LLM-powered answer synthesis with source citations
Scope: Searches across all project knowledge (no agent scoping)

Getting Started

1. Generate an API Key

Navigate to your Qoris project settings
Go to the “API Keys” section
Click “Generate New API Key”
Save the key securely

2. Configure MCP Connection

Add Qoris to your MCP-compatible application configuration:

{
  "mcpServers": {
    "qoris": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-qoris"],
      "env": {
        "QORIS_API_KEY": "your-api-key-here",
        "QORIS_PROJECT_ID": "your-project-id"
      }
    }
  }
}

3. Start Using the Tools

Once configured, you can use any of the 7 tools in your AI application. See individual tool documentation for detailed usage examples.

Use Cases

Conversational AI

Use memory tools to remember user preferences and context across conversations. Search knowledge to answer questions using your project’s documentation.

Automated Workflows

Store workflow state and decisions in memory. Query knowledge base to make informed decisions based on project documentation. Use Knox when the workflow runs commands, calls MCP tools with external input, schedules work, or mutates external systems.

Multi-Agent Systems

Share memories across multiple agents working on the same project. Each agent can save and retrieve context. Knox preserves governance around the full path when subagents perform actions.

Documentation Assistants

Use search_knowledge to create assistants that can answer questions about your project’s documentation, codebase, or knowledge base.

Best Practices

Memory Management

Save important information proactively when you learn it
Keep memory content concise and self-contained
Use agent metadata to track memory sources
Update memories rather than creating duplicates

Knowledge Search

Use specific, detailed queries (2+ words minimum)
Provide context to improve search relevance
Review source citations to verify information
Iterate on queries if initial results aren’t good

Security

Keep API keys secure and never commit them to version control
Use different API keys for different projects
Rotate API keys periodically
Monitor API key usage in project settings
Use Knox for agent execution, MCP-heavy workflows, and autonomous tool use

Error Handling

All tools return consistent error responses:

400 Bad Request: Invalid parameters or validation errors
403 Forbidden: Permission denied (e.g., trying to update another API key’s memory)
404 Not Found: Resource not found (e.g., memory doesn’t exist)
500 Internal Server Error: Server-side errors

See individual tool documentation for specific error codes and messages.

Server Information

Base URL: https://mcp.qoris.ai/mcp
User URL Format: https://mcp.qoris.ai/mcp?api_key=YOUR_API_KEY (each user gets their own URL)
Transport: Streamable HTTP
Authentication: API Key (query param or header) or OAuth 2.0 (Bearer token)
Capabilities: Read + Write
Access Token Expiration: None — OAuth access tokens remain valid until revoked For complete technical details, see our Technical Reference.

Next Steps

Read the Quick Start Guide for setup instructions
Explore individual tool documentation for detailed usage
Review Knox for agent action safety, hooks, policy, and audit
Review the Technical Reference for integration details
Check out API Examples for backend integration

Ready to integrate? Check out our Quick Start Guide, review Knox, or dive into the individual tool documentation.

​MCP Tools Overview

Memory Tools

Knowledge Tools

Knox Governance

​What is MCP?

​Authentication

​Available Tools

​Memory Tools

save_memory

get_memories

update_memory

delete_memory

​Knowledge Tools

search_knowledge

list_knowledge_documents

get_document_full_content

​Data Persistence

​Memories

​Knowledge

​Getting Started

​1. Generate an API Key

​2. Configure MCP Connection

​3. Start Using the Tools

​Use Cases

​Best Practices

Memory Management

Knowledge Search

Security

​Error Handling

​Server Information

​Next Steps

MCP Tools Overview

What is MCP?

Authentication

Available Tools

Memory Tools

Knowledge Tools

Data Persistence

Memories

Knowledge

Getting Started

1. Generate an API Key

2. Configure MCP Connection

3. Start Using the Tools

Use Cases

Best Practices

Error Handling

Server Information

Next Steps