Downloads · 30 days
0
Lukynnnn/mcp-database-universal
mcp-database-universal is a machine learning model from Lukynnnn. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mcp-database-universal. The card lists the license as mit.
A reasoning interface for databases — not a thin execution wrapper. Designed so LLM agents understand what the data means, not just how to fetch it.
Downloads · 30 days
0
Access
Public
Updated Aug 27, 2026
Repo size
—
Likes
0
Public
Click a slice to open those files.
.py77 KB · 92%
From the Hugging Face model README
A reasoning interface for databases — not a thin execution wrapper. Designed so LLM agents understand what the data means, not just how to fetch it.
test_connection, list_tables, inspect_table, query, natural_query, profile_database, schema_graphDATABASE_WRITE_ENABLED=trueVARCHAR(255) → "text, max 255 chars"), NULL → (empty), output size capDATABASE_URL=sqlite:///path/to.dbpip install mcp-database-universal
pip install "mcp-database-universal[postgres]" # PostgreSQL support
pip install "mcp-database-universal[mysql]" # MySQL support
pip install "mcp-database-universal[mssql]" # MSSQL support
DATABASE_URL=sqlite:///app.db python -m mcp_database_universal
docker run -i -e DATABASE_URL=sqlite:////data/app.db \
-v ./data:/data mcp-database-universal:latest
| Variable | Description | Default |
|---|---|---|
DATABASE_URL | sqlite:///path.db, postgresql://..., mysql://..., mssql://... | required |
DATABASE_READ_ONLY | Enforce read-only mode | true |
DATABASE_WRITE_ENABLED | Allow write operations (INSERT/UPDATE/DELETE) | false |
DATABASE_MAX_ROWS | Max rows returned per query | 1000 |
DATABASE_MAX_QUERY_TIME | Query timeout in seconds | 30 |
DATABASE_MAX_OUTPUT_BYTES | Cap on result payload size | 50000 |
OPENAI_API_KEY | API key for the LLM-backed natural_query (OpenAI) | — |
ANTHROPIC_API_KEY | API key for the LLM-backed natural_query (Anthropic) | — |
| Tool | Description |
|---|---|
test_connection | Verify connection, get engine type, version, size |
list_tables | Overview of tables with row counts and relationships |
inspect_table | Full table structure: columns, types, keys, sample data |
query | Execute parametrized, safety-checked SQL |
natural_query | Ask in plain language, get SQL + results + explanation |
profile_database | Value distributions, NULL rates, relationships, sizes |
schema_graph | Mermaid ER diagram of relationships |
MIT