Manticore Search MCP
OfficialThe Manticore Search MCP server enables AI assistants to query, explore, and manage Manticore Search databases through a standardized interface.
Execute SQL queries (
run_query): Run SQL statements (SELECT, SHOW, DESCRIBE, etc.) against a Manticore Search instance. Write operations (INSERT, UPDATE, DELETE) and destructive operations (DROP, TRUNCATE) are disabled by default and require explicit configuration via environment variables.List tables/indexes (
list_tables): Retrieve all available tables and indexes in the connected Manticore Search database.Describe table schema (
describe_table): Get detailed schema information for a specific table or index, including column names and types.List documentation (
list_documentation): Browse and filter available Manticore Search manual files using regex patterns.Fetch documentation content (
get_documentation): Retrieve full or filtered content from a specific manual page (e.g., full-text search operators, KNN vector search, data types), with optional context lines around matched sections.Built-in LLM guidance (
manticore_initial_prompt): Provides AI assistants with knowledge about Manticore Search features to improve query generation.Health monitoring: HTTP endpoint at
/healthfor connectivity checking.Flexible deployment: Supports local stdio and remote HTTP transport with authentication, compatible with Claude, Cursor, Windsurf, VS Code, and more.
Provides tools to list and fetch specific documentation from the Manticore Search manual, utilizing the GitHub API for content retrieval.
Manticore Search MCP Server
MCP server for Manticore Search — enables AI assistants to query and manage Manticore Search databases directly.
Quick Start
Installation
# Option 1: Install with uv (recommended, requires PyPI release)
uvx mcp-manticore
# Option 2: Install with pip
pip install mcp-manticore
# Option 3: Run from source (for local development)
uvx --from . mcp-manticore
# Or: uv run mcp-manticoreNote:
uvxruns the package directly without installation. First-time run may take a moment to download dependencies.The package must be published to PyPI for
uvx mcp-manticoreto work.For local development or testing unreleased versions, use
uvx --from . mcp-manticore
Related MCP server: MCP MariaDB Server
What It Does
Tools
Tool | Description |
| Execute SQL queries (SELECT, SHOW, DESCRIBE, etc.) |
| List all tables and indexes |
| Get table schema |
| List available documentation files |
| Fetch specific documentation from Manticore manual |
Prompts
manticore_initial_prompt— Built-in prompt teaching LLMs about Manticore Search features (full-text operators, KNN vector search, fuzzy search, etc.)
Health Check
HTTP endpoint at /health for monitoring connectivity.
Configuration
Environment Variables
Variable | Default | Description |
|
| Manticore server host |
|
| HTTP API port |
| — | Username (optional) |
| — | Password (optional) |
|
| Connection timeout (seconds) |
|
| Query timeout (seconds) |
|
| Enable write operations (INSERT, UPDATE, DELETE) |
|
| Enable destructive operations (DROP, TRUNCATE) |
| — | GitHub token for higher API rate limit |
Safety
By default, all write operations are blocked. To enable:
# Enable writes (INSERT, UPDATE, DELETE)
export MANTICORE_ALLOW_WRITE_ACCESS=true
# Enable destructive operations (DROP, TRUNCATE)
export MANTICORE_ALLOW_DROP=trueConnect to Your AI Assistant
Open terminal and run:
claude mcp add manticore -- uvx mcp-manticoreOr with environment variables:
claude mcp add manticore -- uvx mcp-manticore -- \
MANTICORE_HOST=localhost \
MANTICORE_PORT=9308For full configuration, edit ~/.claude/mcp_settings.json:
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Restart Claude Code or type /mcp restart to apply changes.
Method 1: Via Settings UI
Open Cursor → Settings → Tools & MCP
Click "Add MCP Server"
Enter name:
manticoreCommand:
uvx mcp-manticore
Method 2: Via Config File
Global config (~/.cursor/mcp.json):
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Project config (.cursor/mcp.json in your project):
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"]
}
}
}Method 1: Via Cascade UI
Open Windsurf → Cascade panel
Click the MCPs icon (🔨) in the top-right
Click "Add Server"
Enter:
uvx mcp-manticore
Method 2: Via Config File
Edit ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Or open directly in Windsurf: Cmd/Ctrl + Shift + P → "MCP Configuration Panel"
Run in terminal:
codex mcp add manticore -- uvx mcp-manticoreWith environment variables:
codex mcp add manticore \
--env MANTICORE_HOST=localhost \
--env MANTICORE_PORT=9308 \
-- uvx mcp-manticoreOr edit ~/.codex/config.toml directly:
[mcp_servers.manticore]
command = "uvx"
args = ["mcp-manticore"]
env = { MANTICORE_HOST = "localhost", MANTICORE_PORT = "9308" }Edit ~/.gemini/settings.json:
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Create .vscode/mcp.json in your workspace:
{
"servers": {
"manticore": {
"type": "stdio",
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Open Cline panel in VS Code
Click the MCP Servers icon
Click "Configure" → "Add Server"
Select "Command (stdio)" and enter:
Name:
manticoreCommand:
uvx mcp-manticore
Or edit the MCP settings file directly (accessible via the MCP Servers icon → "Edit Config"):
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Open Roo Code panel in VS Code
Click the MCP Servers icon → "Edit MCP Settings"
Add the server configuration
Or create .roo/mcp.json in your project root:
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%/Claude/claude_desktop_config.json (Windows):
{
"mcpServers": {
"manticore": {
"command": "uvx",
"args": ["mcp-manticore"],
"env": {
"MANTICORE_HOST": "localhost",
"MANTICORE_PORT": "9308"
}
}
}
}HTTP Transport (Remote MCP)
By default, MCP uses stdio (local). For remote access:
export MANTICORE_MCP_SERVER_TRANSPORT=http
export MANTICORE_MCP_BIND_PORT=8000
export MANTICORE_MCP_AUTH_TOKEN="your-secure-token"
uvx mcp-manticoreConnect via URL:
{
"mcpServers": {
"manticore": {
"url": "http://localhost:8000/mcp",
"headers": {
"Authorization": "Bearer your-secure-token"
}
}
}
}Troubleshooting
Install uv (required)
macOS / Linux:
# Via installer script (recommended)
curl -LsSf https://astral.sh/uv/install.sh | sh
# Or via Homebrew
brew install uvWindows:
# Via PowerShell
irm https://astral.sh/uv/install.ps1 | iex
# Or via winget
winget install astral-sh.uvVerify installation:
uv --versionMCP server not connecting
Verify Manticore is running:
curl http://localhost:9308/healthCheck environment variables are set correctly
For Claude Code: restart with
/mcp restart
Too many tools loaded
Some agents limit active MCP tools. Remove unused servers or use project-scoped configs.
Development
# Clone and setup
git clone https://github.com/manticoresoftware/mcp-manticore.git
cd mcp-manticore
# Install dependencies
uv sync
# Run locally
uv run mcp-manticore
# Run with custom config
MANTICORE_HOST=remote-server MANTICORE_PORT=9308 uv run mcp-manticore
# Run tests
uv run pytest
# Build package
uv build
# Publish to PyPI
uv publishArchitecture
File | Purpose |
| Configuration management |
| MCP server implementation |
| LLM guidance/prompts |
| GitHub docs fetcher |
| CLI entry point |
License
Available Tools
5 toolsdescribe_tableA
Get the schema of a specific table/index in Manticore Search.
Args: table_name: The name of the table/index to describe
Returns: A dictionary containing column information including: - columns: List of column names and types
Inputs: table_name (required). Returns: object
get_documentationA
Fetch documentation from Manticore Search manual.
Use list_documentation() first to discover available files.
Args: file_path: Path to documentation file (e.g., "Searching/KNN.md") content: Optional search term to filter content (returns only matching sections) before: Number of lines before match to include (default: 0) after: Number of lines after match to include (default: 0)
Returns: Documentation content as markdown text
Examples: get_documentation( "Searching/Full_text_matching/Operators.md", content="MATCH", before=2, after=2 )
get_documentation("Creating_a_table/Data_types.md")Inputs: file_path (required), content, before, after. Returns: result
list_documentationA
List available documentation files from Manticore Search manual.
IMPORTANT: Make ONE call with regex OR patterns to find multiple topics at once. DO NOT make multiple calls - use pipe (|) to combine patterns.
The file list is cached, so filtering is instant and local.
Args: search: Regex pattern to filter files (case-insensitive). Use OR patterns (|) to match multiple topics in ONE call:
EFFICIENT (single call):
- "knn|vector|embedding" - find all vector-related docs
- "search|query|match" - find all search-related docs
- "^Searching/|^Creating_a_table/" - multiple directories
- "index|table|cluster" - multiple topics
INEFFICIENT (avoid multiple calls):
- Don't: list_documentation("knn"), then list_documentation("vector")
- Do: list_documentation("knn|vector") # ONE callReturns: Filtered list of documentation files, grouped by category
Examples: # List all documentation (no filter) list_documentation()
# Find multiple topics in ONE call (RECOMMENDED)
list_documentation(search="knn|vector|embedding")
list_documentation(search="search|query|match|full-text")
list_documentation(search="index|table|cluster|replication")
# Directory-specific searches
list_documentation(search="^Searching/") # all in Searching/
list_documentation(search="^Creating_a_table/") # all in Creating_a_table/
# Combine directories and topics
list_documentation(search="^Searching/|^Creating_a_table/")Inputs: search. Returns: result
list_tablesA
List available tables/indexes in Manticore Search.
Returns: JSON string containing list of table names and their types
Inputs: no named parameters. Returns: result
run_queryA
Execute a SQL query against Manticore Search.
Queries run in read-only mode by default. Set MANTICORE_ALLOW_WRITE_ACCESS=true to allow DDL and DML statements when your Manticore server permits them.
Args: query: The SQL query to execute (e.g., "SELECT * FROM my_index LIMIT 10")
Returns: A dictionary containing: - columns: List of column names - rows: List of row values - total: Total number of results
Inputs: query (required). Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool has a distinct and non-overlapping purpose: describe_table for schema inspection, list_tables for inventory, run_query for data retrieval/manipulation, and two documentation tools (list_documentation and get_documentation) that work together as a clear pair. No ambiguity exists between tools.
All tools follow a consistent verb_noun naming pattern (describe_table, get_documentation, list_documentation, list_tables, run_query). The verbs are appropriate and predictable, with no mixing of conventions or styles.
Five tools is well-scoped for a search database server, covering core operations: schema inspection, table listing, query execution, and documentation access. Each tool serves a clear purpose without redundancy or bloat.
The toolset covers essential read operations and documentation access effectively. A minor gap exists in write operations (e.g., create/update/delete tables or documents), though run_query hints at DDL/DML capabilities with server permissions. For a search-focused server, this is reasonable but not fully comprehensive.
Maintenance
Related MCP Connectors
Query Postgres, MySQL, SQL Server, Oracle, BigQuery, ClickHouse and Redshift from your AI client.
Store, query, and update structured data from any AI agent
The all-in-one data stack for agents. Upload files, run SQL, evolve tables, and render charts.
PostgreSQL, MySQL, and SQL Server in one session. 26 read-only MCP tools for AI agents.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to interact with MariaDB databases through standard SQL operations and advanced vector/embedding-based search. Supports database management, schema inspection, and semantic document storage and retrieval with multiple embedding providers.MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to interact with MariaDB databases through SQL operations and vector-based semantic search. Supports standard database queries, schema inspection, and optional embedding-powered document storage and retrieval.MIT
- AlicenseBqualityDmaintenanceEnables AI assistants to securely interact with MariaDB and MySQL databases using granular per-connection read/write permissions and transaction support. It allows users to manage multiple database connections, explore schemas, and execute controlled SQL queries through a standardized interface.617 npm5MIT
- AlicenseNot gradedqualityCmaintenanceConnect AI agents to SQL databases (SQLite, PostgreSQL, MySQL) with a unified interface for querying data, exploring schemas, inserting rows, and exporting results to CSV. Includes safety features like dangerous query blocking and write guards.2 npm39 PyPIMIT