MongoDB.

Connect Orchid to MongoDB Atlas, self-hosted MongoDB, or a DocumentDB-compatible deployment. Cells use MongoDB's native query syntax — find, aggregation pipelines, and a small set of helpers.

Requirements

• MongoDB 5.0+ recommended.
• A connection string from your provider (Atlas, self-hosted, etc.).
• A database user with at least read on the databases you want to query.

Connect

1. Open Integrations → + Add connection → MongoDB.
2. Paste your connection string, or fill in the fields manually:
   • Connection string — e.g. mongodb+srv://user:pass@cluster.mongodb.net/mydb?retryWrites=true&w=majority
   • Database — the default database to query against
3. Click Test connection. Green check = ready.

Connection string format

MongoDB uses URI-style connection strings. Two common shapes:

# Standard format (self-hosted, single host or replica set)
mongodb://username:password@host1:27017,host2:27017,host3:27017/mydatabase?replicaSet=rs0&authSource=admin

# DNS-seedlist format (Atlas, simpler)
mongodb+srv://username:password@cluster0.mongodb.net/mydatabase?retryWrites=true&w=majority
      

Useful URI parameters:

• authSource — which database holds the user's credentials. Almost always admin.
• replicaSet — for replica-set discovery.
• tls=true — required for Atlas.
• retryWrites=true — retries idempotent writes after a transient failure.
• w=majority — write concern; default in Atlas.

Optional settings

Atlas IP allow list

Atlas blocks connections from un-allow-listed IPs. Add your current IP in Network Access → Add IP address. For a long-lived setup, allow a range from your office or VPN. For ad-hoc work, click Add Current IP Address from the Atlas UI.

SSH tunnel

Self-hosted MongoDB on a private VPC works through an SSH tunnel like any other connector. Bastion host, user, private key.

Writing queries

MongoDB cells use MongoDB shell syntax. Orchid wraps it in a small evaluation surface that returns documents to the result table.

Simple find

db.users.find(
  { status: "active", created_at: { $gt: ISODate("2026-04-01") } },
  { name: 1, email: 1, last_login: 1 }
).limit(50)
      

Aggregation pipeline

db.orders.aggregate([
  { $match: { created_at: { $gt: ISODate("2026-04-01") } } },
  { $group: {
      _id: { day: { $dateTrunc: { date: "$created_at", unit: "day" } } },
      orders: { $sum: 1 },
      revenue: { $sum: "$total" }
  }},
  { $sort: { "_id.day": 1 } }
])
      

Lookup (join)

db.orders.aggregate([
  { $match: { status: "shipped" } },
  { $lookup: {
      from: "users",
      localField: "user_id",
      foreignField: "_id",
      as: "user"
  }},
  { $unwind: "$user" },
  { $project: { _id: 1, total: 1, "user.email": 1 } }
])
      

Common gotchas

• "MongoServerError: bad auth" — wrong username/password, or wrong authSource. Almost always fixed by adding ?authSource=admin to the URI.
• "MongooseServerSelectionError: connection timed out" — Atlas IP not allow-listed, or self-hosted server not reachable. Check your IP allow list first.
• Slow first query — Atlas free-tier clusters pause after inactivity. First query after a pause takes ~5 seconds while the cluster resumes.
• "Field path references must be prefixed with a dollar sign" — you wrote $group: { _id: "user_id" } (string literal) instead of $group: { _id: "$user_id" } (field reference). MongoDB takes the dollar prefix seriously.
• Numbers come back as NumberLong / NumberDecimal — Mongo's extended JSON types. Orchid converts these to regular numbers in the result table, but if you export raw JSON the type wrappers come back. Use $toLong / $toDouble in your pipeline to coerce.
• DocumentDB / Azure Cosmos compatibility — both speak the MongoDB wire protocol but with feature gaps (no $lookup on DocumentDB before 4.0, no transactions in some Cosmos tiers). Stick to documented features for portability.

Example queries

// List collections in the current database
db.getCollectionNames()
      

// Count documents per status
db.orders.aggregate([
  { $group: { _id: "$status", count: { $sum: 1 } } },
  { $sort: { count: -1 } }
])
      

Where to go next

For more on writing SQL cells, see SQL cells.