termalData · MongoDB
A MongoDB GUI that tunnels to the database for you.
Browse collections, filter with a JSON query, run an aggregation pipeline, and edit a document by its _id — inside a full SSH client. Reach a MongoDB that only listens on the server through an SSH tunnel that opens itself, or a private one through a keyless Connector. A Compass-style workflow without a separate tunnel or a second app.
Free tier · 14-day Pro trial · Windows, macOS & Linux — or the web cabinet
01 / find and aggregate
Query documents in the JSON you already know.
Open a collection and Termalin lists its documents, flattening them into a grid whose columns are the union of top-level fields, _id first. Type a JSON filter to narrow the set, or a JSON array to run a full aggregation pipeline. Because Mongo has no fixed schema, it samples documents to show you what fields a collection actually holds.
- ▸Filter or aggregate —
{"status":"paid"}to find,[…]to run a pipeline. - ▸Fields inferred — a sampled field list stands in for the schema Mongo doesn't have.
- ▸Collections and views — with estimated document counts, browsed like any host.
02 / edit by _id
Change one field, not the whole document.
Edit a cell in the grid and Termalin addresses that document by its _id with an updateOne — wrapped as a partial $set, so only the field you touched changes and the rest of the document is left alone. Add a document from a JSON form (an insertOne), delete one by _id, and create or drop a collection by name. Every command it builds is shown before it runs, so nothing executes unseen.
03 / reach it safely
Through an SSH host or a Connector — never exposed.
A production database shouldn't be on the public internet. On the desktop, connecting to a MongoDB bound to the server opens an SSH local forward through your host automatically — the manual version is in our guide to database tunnels. For a network with no inbound port, a one-line Connector reaches in over an outbound link. TLS is enabled for connections that need it, so AWS DocumentDB — same wire protocol as MongoDB — works too.
04 / and twelve more engines
Not just Mongo — one client for your whole stack.
MongoDB is one of thirteen engines in the same client, next to Postgres, MySQL, SQLite, SQL Server, Redis, ClickHouse, Elasticsearch, Cassandra, Neo4j, InfluxDB, RabbitMQ and Cloudflare D1. Your connections live in the same encrypted vault as your SSH hosts and keys, and an AI agent can read them read-only by default. See the full database client →
Questions
Can this MongoDB GUI connect over an SSH tunnel?
Yes — on the desktop, connect to a MongoDB that only listens on the server and Termalin opens an SSH local forward through your host for you, no ssh -L to wire up. For a private network with no inbound SSH port, reach it through a keyless Connector. TLS is switched on automatically for connections that require it, including AWS DocumentDB.
Is it a MongoDB Compass alternative?
It does the everyday Compass jobs — browse collections, filter with a JSON query, run an aggregation, edit and add documents — but inside a full SSH client, and it reaches a database through an SSH host or a Connector without a separate tunnel. It also speaks twelve other engines, so it isn't one tool per database.
How do I browse and query a collection?
Open a collection and Termalin runs collection {} to list documents, flattening them into a grid whose columns are the union of top-level fields, _id first. Type a JSON filter — orders {"status": "paid"} — to narrow it, or a JSON array to run an aggregation pipeline. It samples documents to show a collection's fields even though Mongo has no fixed schema.
Can I edit and add documents?
Yes — edit a cell and Termalin addresses the document by its _id with an updateOne (a partial $set), so only that field changes. Add a document from a JSON form (insertOne), delete one by _id, and create or drop a collection by name. Every generated command is shown before it runs.
Can an AI agent read MongoDB safely?
Yes — the built-in MCP server exposes read tools that are read-only by default and per-connection, so an agent can answer questions about a collection without write access or the credential. More in giving an agent database access.
Your MongoDB, in the app you already keep open.
Browse collections, filter and aggregate in JSON, edit documents by _id — reaching a private database through an SSH tunnel or a Connector, in the app and the browser. Free to start; new accounts get Pro free for 14 days.
Free tier · 14-day Pro trial · Redis GUI · the Connector