Home
DBMojo Core is the platform underneath five independently-installable tools — Detanami, DataMosaic, Flow, Smith, and Blueprint — providing single sign-on, shared data connections, approval-gated deployments, version control, and tamper-evident auditing across all of them.
My Suite
DetanamiConnect, browse, and tune SQL and NoSQL databases from one console.
DataMosaicBuild dashboards from a visual query graph and share them live with your team.
FlowDrag-and-drop orchestration across databases, APIs, and message queues.
SmithAn AI co-pilot that writes queries and generates deployable data APIs.
BlueprintA web-based data modeler that reverse-engineers a schema and files changes as approved deployments.
Platform
The platform every tool runs on — one identity, one credential vault, one approval ledger, one audit trail, shared across Detanami, DataMosaic, Flow, Smith, and Blueprint. None of the five tools run without it.
Core isn't a proxy in front of the tools — it's the identity, credential, approval, and audit layer they all read from. It never connects to your databases itself; the tool that owns a connection does the work.
Review before approving: the approver sees the exact SQL, the requester's note, and a comment thread — here, an index suggested by Detanami's Analyze & Tune.
Bring your own model. One page sets the AI backend for every tool — self-hosted Ollama or vLLM for air-gapped installs, or OpenAI, Azure OpenAI, Anthropic, Bedrock, or Gemini.
Audit you can verify. A hash-chained log you can re-check on demand, with compliance-named retention presets and PHI/PII scrubbing.
Database management
DetanamiA web-based client for connecting to, exploring, and managing your databases — SQL and NoSQL alike — without switching tools per engine.
Detanami is designed as an intuitive NoSQL client at heart, extended to cover the SQL engines teams already run in production. Redis and MongoDB get their own dedicated consoles instead of being forced through a SQL workbench that doesn't fit them.
Analyze & Tune on a live PostgreSQL table: the plan finding (a sequential scan on beerreview_beer), a plain-language explanation, and the suggested CREATE INDEX — ready to file for approval.
Visualization
DataMosaicConnect a data source, drag out a dashboard, and share it — built for the back-and-forth between analysts and the people asking them questions.
The visual query graph behind a Top 10 by State widget: Table → Aggregate → Sort → Limit, compiled to SQL and previewed live.
The finished BeerSchooner dashboard — a table, a bar chart, and a map, all backed by queries against the same shared connection.
Orchestration
FlowInspired by Node-RED — wire up sources, transforms, and destinations visually, and let Flow run the pipeline on schedule or in response to events.
A deployed fan-out/fan-in flow — two database branches rejoin in a Join node — with the Toolbox collapsed to show the Network nodes, including Kafka In and Kafka Out.
AI-assisted development
SmithA notebook where an AI agent reasons about your schema, not just your prompt — then generates a real, callable data API you can preview on the spot and deploy once it's approved.
Smith bridges AI reasoning and database physics: it writes SQL and Node.js against your actual schema, generates a deterministic Data API from any table, previews it locally against the real database, and can serve it from API workers in your own Docker deployment, ship it to AWS Lambda, or save it to your Git repo — with Lambda and API-worker deploys of Core connections going through Core's approval ledger and the generated code in front of the approver.
A live notebook: tabbed cells on the left, query results below, and the AI Assistant docked on the right — it answers with runnable SQL you choose to insert.
Data modeling
BlueprintA web-based data modeler: reverse-engineer a live schema onto an ER canvas, edit it, and turn the difference into a reviewed, approved change — without Blueprint ever touching your database.
Blueprint reads a schema from a Shared Data Connection — credentials are resolved from Core on each request and never stored — and draws it as tables and relationships. You model changes on the canvas; Blueprint generates the DDL, compares it with the live database, and files the result with a rollback script as a request in Core. It never runs DDL itself: an approver reads the exact script, and Detanami executes it.
A MySQL schema imported through a Shared Data Connection. Each line carries its cardinality (0..* to 1), the selected relationship is explained in plain words in the inspector, and the legend decodes the marks.
Coverage
Detanami connects to the engines already in your stack — no separate client per database type.
All support TLS/SSL, except SQLite and DuckDB (local files have no network connection to secure) and Firebird; the cloud warehouses are always TLS.
Add Detanami, DataMosaic, Flow, Smith, and Blueprint as your workflow needs them; skip the ones you don't.
DBMojo