upjack 0.0.1

Schema-driven entity management for AI-native applications

upjack

Schema-driven entity management for AI-native applications.

Define entities as JSON Schemas + a manifest. The library handles CRUD, validation, ID generation, storage, and search. Add FastMCP to serve it as an MCP server.

Install

From PyPI (once published):

# Core library (entity management only)
uv add upjack

# With MCP server support
uv add upjack[mcp]

From source (alpha):

git clone https://github.com/NimbleBrainInc/upjack.git
cd upjack/lib/python
uv pip install -e ".[mcp]"

Usage: Entity Management

from upjack import UpjackApp

app = UpjackApp.from_manifest("manifest.json")

# Create an entity
contact = app.create_entity("contact", {
    "first_name": "Sarah",
    "last_name": "Chen",
    "email": "[e-mail]",
})

# List entities
contacts = app.list_entities("contact")

# Update an entity
app.update_entity("contact", contact["id"], {"lead_score": 85})

# Get a single entity
contact = app.get_entity("contact", contact["id"])

# Delete an entity (soft delete by default)
app.delete_entity("contact", contact["id"])

Usage: MCP Server

from upjack.server import create_server

mcp = create_server("manifest.json", root="./workspace")
mcp.run()

This generates CRUD tools for every entity type in the manifest, exposes context and skills as MCP resources, and handles validation automatically. See examples/crm/server.py for a complete example.

Requirements

• Python >= 3.13
• FastMCP >= 3.0 (only for upjack[mcp])