Provides read-only PostgreSQL schema metadata introspection, including listing schemas, tables, and columns, describing table details, retrieving foreign key relationships, exporting schema documentation, and refreshing the schema cache.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SchemaPilot MCP describe the users table"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for PostgreSQL schema metadata — read-only, local-first, AI-coding-agent ready.
Read-only PostgreSQL introspection (no writes, no arbitrary SQL)
MCP protocol (Model Context Protocol) for AI coding agents
Schema/table/column metadata with PK/FK/index details
Explicit cache refresh (sync_schema)
Markdown documentation export (deterministic)
Schema allowlist + max table/column limits
Secret redaction in logs
Related MCP server: supabase-schema-mcp
pip install -e .
# Configure
cp schemapilot.yaml.example schemapilot.yaml
# Edit with your PostgreSQL credentials
schemapilotTool | Description |
| List available schemas |
| List tables with column counts |
| Full table details |
| FK relationship graph |
| Markdown documentation |
| Refresh schema cache |
postgres:
host: localhost
port: 5432
database: mydb
user: app
password: secret
schema_allowlist:
- public
- app
max_tables: 100
max_columns: 500
cache_ttl_seconds: 300MIT
This server cannot be deployed
PostgreSQL, MySQL, and SQL Server in one session. 26 read-only MCP tools for AI agents.
Analytical memory for AI agents: a real Postgres queried in plain English over MCP. One command.
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.