PLOMID Studio
The database workbench for PLOMID.
A desktop workspace for a PLOMID server: browse objects, read and edit data, work with documents, vectors and time-ordered rows, and run SQL — without a hosted console sitting between you and the engine.
-- Studio opens a connection before it reads anything
SELECT seat, state, since
FROM cabin_states
WHERE vehicle_id = :vin
ORDER BY since DESC; No rows. Studio states what it has not read rather than inventing data.
Where Studio is today.
Studio is an engineering preview. The application shell, the workspace surfaces, the design system, the domain models and the protocol codec are built and tested — but the transport that reads a server is not connected yet. Until it is, every surface states what it has not read instead of showing invented data.
- Built as
- A desktop application: React and TypeScript in the renderer, a thin native shell.
- Talks to
- The engine’s PostgreSQL-compatible endpoint for data, the Admin API for management.
- Native for
- Only what the operating system owns: secure credential storage, file dialogs, notifications, updates.
- State today
- Shell, surfaces, design system, domain models and protocol codec are in place with tests.
What the workspace is for.
Each surface answers one job. They share the same connection, the same object model and the same command registry, so a table found in the browser is one keystroke from the workbench.
- 01 Object browser Connections, databases, schemas, tables and their children as one navigable tree, with search and keyboard movement.
- 02 Object inspector The selected object in full: columns and types, constraints, indexes, and what the engine reports about it.
- 03 Data explorer Rows read as a grid — sorting, filtering, paging, and an export path — against the live database.
- 04 Document view JSON and document values parsed into a tree with paths, so a nested field is edited in place rather than as a string.
- 05 SQL workbench Editor, formatting, completion, history and snippets, with the result set and the engine message beneath the statement.
- 06 Vector and time-series workspaces Purpose-built views for the two models a generic grid serves badly: embeddings and time-ordered reads.
- 07 Command palette One registry for every action the application can perform, reachable without leaving the keyboard.
How Studio reaches a database.
Studio is a client, never a server. Two documented boundaries carry everything it does — one for data, one for management — and nothing else in the application talks to the engine behind those boundaries.
- 01 Workspace Objects, data, documents and queries in one desktop application.
- 02 pgwire The engine’s PostgreSQL-compatible wire protocol, used for data.
- 03 PLOMID The system of record. Planning, execution and storage stay in the engine.
- Database developers working directly against a PLOMID server.
- Teams that run the database where the data lives and want the tool beside it.
- Anyone evaluating PLOMID, reading real objects rather than screenshots.
Studio is an engineering preview. The application shell, the workspace surfaces and the design system are in place with tests, but there is no server transport yet: every surface states what it has not read instead of showing invented data.
Build against the engine today.
The database Studio connects to is available now. Install it, run a first query, and follow Studio as the transport lands.