About sqlite-reader-mcp
sqlite-reader-mcp is a read-only MCP server for inspecting SQLite databases. It is intended for clients that need to query data and examine schemas while restricting access to approved paths and read queries.
Key features
- 01Run SELECT and WITH queries
- 02List database tables
- 03Describe table schemas
- 04Restrict access to approved paths
- 05Limit query results
Use cases
- 01Inspect rows in an approved SQLite database
- 02Check table names before querying
- 03Review column types and constraints
- 04Query data without modifying the database
MCP support
The MCP server provides tools to run read queries, list tables, and describe schemas in approved SQLite databases. It rejects statements other than SELECT or WITH queries.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEsqlite-reader-mcp
MCP server for reading SQLite databases.
Overview
sqlite-reader-mcp is a lightweight MCP server that provides read-only access to SQLite databases. It allows users to execute SELECT queries, list tables, and describe table schemas.
Features
- Read-Only Access: Ensures data integrity by only allowing read operations (SELECT queries).
- Secure Path Handling: Restricts database access to pre-approved absolute file paths and directories.
- SQL Validation: Parses and validates SQL queries, permitting only
SELECTandWITHstatements to prevent unintended operations. - Asynchronous Operations: Utilizes
aiosqlitefor non-blocking database interactions. - Row Limiting: Automatically limits the number of rows returned by queries to prevent excessive data retrieval.
- MCP Tools:
read_query: Execute a SELECT SQL query on a specified SQLite database.list_tables: List all tables within a given SQLite database.describe_table: Get schema information (column names, types, etc.) for a specific table.
Requirements
- Python 3.13+
aiosqlite >= 0.21.0fastmcp >= 2.6.1sqlparse >= 0.5.3
Installation
The recommended installation method is with uv:
uv tool install git+https://github.com/abhinavnatarajan/sqlite-reader-mcp
Usage
sqlite-reader-mcp --paths /path/to/your/database.db /path/to/another/allowed_directory
Arguments:
--paths(or-p): A list of absolute paths to SQLite database files or directories containing them that the server is allowed to access.
Once the server is running, you can interact with it using a FastMCP client, utilizing the following tools:
read_query
Executes a SELECT query on a specified SQLite database file.
- Args:
file_path(str): Absolute path to the SQLite database file.query(str): The SELECT SQL query to execute.params(Optional[List[Any]]): Optional list of parameters for the query.fetch_all(bool): IfTrue(default), fetches all results. IfFalse, fetches one row.row_limit(int): Maximum number of rows to return (default 1000).
- Returns: A list of dictionaries, where each dictionary represents a row from the query result.
list_tables
Lists all tables in the specified SQLite database file.
- Args:
file_path(str): Absolute path to the SQLite database file.
- Returns: A list of table names (strings).
describe_table
Provides detailed schema information for a specific table in a SQLite database.
- Args:
file_path(str): Absolute path to the SQLite database file.table_name(str): The name of the table to describe.
- Returns: A list of dictionaries, each describing a column (e.g., name, type, notnull, default value, primary key status).
Security
- Read-Only: The server is strictly read-only. No
INSERT,UPDATE,DELETE, or other data modification SQL commands are permitted. - Path Whitelisting: Access is restricted to SQLite files located in paths explicitly provided via the
--pathscommand-line argument. Paths must be absolute and existent. - Query Validation: All SQL queries are parsed. Only
SELECTstatements (including those with Common Table Expressions usingWITH) are allowed. Multiple statements in a single query string are rejected. Trailing semicolons are automatically removed.
Development
This project is built using setuptools. Configuration is in pyproject.toml.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.