Allows interaction with PostgreSQL databases, providing tools for listing schemas, tables, describing table structures, querying relationships, and executing SQL queries.
PostgreSQL MCP Server
A PostgreSQL MCP server implementation using the Model Context Protocol (MCP) Python SDK- an open protocol that enables seamless integration between LLM applications and external data sources. This server allows AI agents to interact with PostgreSQL databases through a standardized interface.
Features
- List database schemas
- List tables within schemas
- Describe table structures
- List table constraints and relationships
- Get foreign key information
- Execute SQL queries
- Typed tools with JSON/markdown output
- Optional table resources and guidance prompts
Quick Start
Installation
Installing via Smithery
To install PostgreSQL MCP Server for Claude Desktop automatically via Smithery:
Manual Installation
- Clone this repository:
- Create and activate a virtual environment (recommended):
- Install dependencies:
Usage
- Start the MCP server.
- The server provides the following tools:
query
: Execute SQL queries against the databaselist_schemas
: List all available schemaslist_tables
: List all tables in a specific schemadescribe_table
: Get detailed information about a table's structureget_foreign_keys
: Get foreign key relationships for a tablefind_relationships
: Discover both explicit and implied relationships for a tabledb_identity
: Show current db/user/host/port, search_path, and version
Typed (preferred):
run_query(input)
: Execute with typed input (sql
,parameters
,row_limit
,format: 'markdown'|'json'
).run_query_json(input)
: Execute and return JSON-serializable rows.list_schemas_json(input)
: List schemas with filters (include_system
,include_temp
,require_usage
,row_limit
).list_schemas_json_page(input)
: Paginated listing with filters andname_like
pattern.list_tables_json(input)
: List tables within a schema with filters (name pattern, case sensitivity, table_types, row_limit).list_tables_json_page(input)
: Paginated tables listing with filters.
Examples:
Inspect current connection identity:
List schemas (JSON) with filters:
Paginated list with pattern filter:
Response shape:
List tables with filters (JSON):
Paginated tables listing:
Resources (if supported by client):
table://{schema}/{table}
for reading table rows. Fallback tools are available:list_table_resources(schema)
→table://...
URIsread_table_resource(schema, table, row_limit)
→ rows JSON
Prompts (registered when supported; also exposed as tools):
write_safe_select
/prompt_write_safe_select_tool
explain_plan_tips
/prompt_explain_plan_tips_tool
Running with Docker
Build the image:
Run the container without a database connection (the server stays inspectable):
Run with a live PostgreSQL database by supplying POSTGRES_CONNECTION_STRING
:
If the environment variable is omitted, the server boots normally and all database‑backed tools return a friendly “connection string is not set” message until you provide it.
Configuration with mcp.json
To integrate this server with MCP-compatible tools (like Cursor), add it to your ~/.cursor/mcp.json
:
Transport Environment Variables
MCP_TRANSPORT=stdio|sse|streamable-http
(default:stdio
)MCP_HOST=0.0.0.0
andMCP_PORT=8000
for SSE/HTTP transportsMCP_SSE_MOUNT=/mcp
optional SSE mount path
If POSTGRES_CONNECTION_STRING
is omitted, the server still starts and is fully inspectable; database‑backed tools will simply return an informative error until the variable is provided.
Replace:
/path/to/venv
with your virtual environment path/path/to/postgres_server.py
with the absolute path to the server script
HTTP Client Integration
Run the server with Streamable HTTP:
Basic reachability check (expect non-200 since MCP expects a handshake):
Example MCP client config (conceptual) pointing at the Streamable HTTP endpoint:
For SSE instead of Streamable HTTP:
Python MCP Client Example (Streamable HTTP)
Security
- Never expose sensitive database credentials in your code
- Use environment variables or secure configuration files for database connection strings
- Consider using connection pooling for better resource management
- Implement proper access controls and user authentication
Environment options
POSTGRES_READONLY=true
to allow only SELECT/CTE/EXPLAIN/SHOW/VALUESPOSTGRES_STATEMENT_TIMEOUT_MS=15000
to cap statement runtime
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Development & Tests
- Create a venv and install runtime deps:
pip install -r requirements.txt
- (Optional) install test deps:
pip install -r dev-requirements.txt
- Run tests:
pytest -q
Related Projects
License
MIT License
Copyright (c) 2025 gldc
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
Enables AI agents to interact with PostgreSQL databases through the Model Context Protocol, providing database schema exploration, table structure inspection, and SQL query execution capabilities.
Related MCP Servers
- -securityFlicense-qualityA server that allows AI models to interact with PostgreSQL databases through a standardized protocol, providing database schema information and SQL query execution capabilities.Last updated -1
- AsecurityAlicenseAqualityEnables AI models to interact with PostgreSQL databases through a standardized interface, supporting operations like queries, table manipulation, and schema inspection.Last updated -63535MIT License
- -securityFlicense-qualityA Model Context Protocol server that provides AI assistants with secure, read-only access to PostgreSQL databases while offering comprehensive tools for schema exploration, query validation, and performance optimization.Last updated -
- AsecurityFlicenseAqualityEnables AI assistants to interact with PostgreSQL databases using natural language queries, providing secure read-only access to database schemas and SQL translation capabilities.Last updated -63