Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/S1LV4/th0th/llms.txt

Use this file to discover all available pages before exploring further.

th0th exposes 8 MCP tools for AI assistants like OpenCode, Claude Desktop, and VSCode extensions. These tools enable semantic code search, persistent memory, context compression, and analytics with 98% token reduction.

​
What is MCP?

Model Context Protocol (MCP) is a standard protocol for connecting AI assistants to external tools and data sources. th0th implements MCP to provide semantic search capabilities directly within your AI coding assistant.

​
Available Tools

th0th_index

Index a project directory for semantic search

th0th_index_status

Check background indexing job progress

th0th_search

Semantic + keyword search with filters

th0th_remember

Store information in persistent memory

th0th_recall

Search stored memories from previous sessions

th0th_compress

Compress context (keeps structure, removes details)

th0th_optimized_context

Search + compress in one call (max token efficiency)

th0th_analytics

Usage patterns, cache performance, metrics

​
Architecture

th0th MCP client connects to the th0th REST API via stdio: 
AI Assistant (OpenCode/Claude)
       ↓ stdio (MCP protocol)
th0th MCP Client
       ↓ HTTP
th0th REST API (port 3333)
       ↓
Semantic Search Engine

​
How it works

  1. MCP Client: Runs as a stdio server, exposing tools to the AI assistant
  2. API Proxy: Proxies tool calls to the REST API via HTTP
  3. REST API: Handles business logic, embeddings, search, and compression
  4. Response Formatting: Returns results in TOON format (70% smaller than JSON)

​
Integration

Via MCP package:
opencode.json
{
  "mcpServers": {
    "th0th": {
      "type": "local",
      "command": ["bunx", "@th0th-ai/mcp-client"],
      "env": {
        "TH0TH_API_URL": "http://localhost:3333"
      },
      "enabled": true
    }
  }
}
Via Plugin:
opencode.json
{
  "plugin": ["@th0th-ai/opencode-plugin"]
}

​
VSCode / Antigravity

Create .vscode/mcp.json in your workspace:
.vscode/mcp.json
{
  "servers": {
    "th0th": {
      "command": "bunx",
      "args": ["@th0th-ai/mcp-client"],
      "env": {
        "TH0TH_API_URL": "http://localhost:3333"
      }
    }
  }
}
Or run the setup script: 
./scripts/setup-vscode.sh

​
Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS): 
{
  "mcpServers": {
    "th0th": {
      "command": "bunx",
      "args": ["@th0th-ai/mcp-client"],
      "env": {
        "TH0TH_API_URL": "http://localhost:3333"
      }
    }
  }
}

​
Docker

{
  "mcpServers": {
    "th0th": {
      "type": "local",
      "command": ["docker", "compose", "run", "--rm", "-i", "mcp"],
      "enabled": true
    }
  }
}

​
Configuration

Configuration file: ~/.config/th0th/config.json (auto-created on first run)

​
Quick config commands

# Show current configuration
npx @th0th-ai/mcp-client --config-show

# Show config file path
npx @th0th-ai/mcp-client --config-path

# Initialize configuration
npx @th0th-ai/mcp-client --config-init

# Show help
npx @th0th-ai/mcp-client --help

​
Embedding providers

ProviderModelCostQuality
Ollama (default)nomic-embed-text, bge-m3FreeGood
Mistralmistral-embed, codestral-embed$$Great
OpenAItext-embedding-3-small$$Great

​
Advanced configuration

# Initialize with specific provider
npx @th0th-ai/mcp-client --config-init                          # Ollama (default)
npx @th0th-ai/mcp-client --config-init --mistral your-api-key   # Mistral
npx @th0th-ai/mcp-client --config-init --openai your-api-key    # OpenAI

# Switch provider
npx @th0th-ai/mcp-client --config-init --mistral your-api-key
npx @th0th-ai/mcp-client --config-init --ollama-model bge-m3

# Set specific configuration values
npx @th0th-ai/mcp-client --config-set embedding.dimensions 1024

​
Next steps

Tools Reference

Detailed documentation for all 8 MCP tools

REST API

HTTP API documentation

Build docs developers (and LLMs) love

Get started for freeTalk to us