Database

Truss wraps PostgreSQL with a SQL workbench, schema browser, ERD visualizer, and a client API for your applications. The dashboard lets you browse tables, run queries, and visualize your schema. The client API gives you SQL-over-HTTP and automatic CRUD endpoints.

SQL-over-HTTP

Run arbitrary SQL via the client API. Requires a service_role API key.

curl -X POST http://localhost:8787/v1/sql \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "sql": "SELECT id, name, email FROM users WHERE active = true",
    "params": [],
    "row_limit": 100
  }'

const res = await fetch('http://localhost:8787/v1/sql', {
  method: 'POST',
  headers: {
    'apikey': 'truss_sk_your_key',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    sql: 'SELECT * FROM users WHERE id = $1',
    params: [42]
  })
});
const { rows, columns, rowCount } = await res.json();

Response includes rows, rowCount, columns (with type info), and command (SELECT, INSERT, etc.).

Transactions

Bundle multiple statements in a single transaction:

curl -X POST http://localhost:8787/v1/sql/transaction \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "statements": [
      {"sql": "INSERT INTO orders (user_id, total) VALUES ($1, $2)", "params": [1, 99.99]},
      {"sql": "UPDATE users SET order_count = order_count + 1 WHERE id = $1", "params": [1]}
    ]
  }'

Up to 20 statements per transaction. If any statement fails, the entire transaction rolls back.

Auto-REST (CRUD)

Every table gets automatic REST endpoints at /v1/db/:table. Works with both anon and service_role keys.

Select rows

# Get all users
curl "http://localhost:8787/v1/db/users" \
  -H "apikey: truss_pk_your_anon_key"

# With filters, ordering, pagination
curl "http://localhost:8787/v1/db/users?active=eq.true&order=created_at.desc&limit=10&offset=0" \
  -H "apikey: truss_pk_your_anon_key"

# Select specific columns
curl "http://localhost:8787/v1/db/users?select=id,name,email" \
  -H "apikey: truss_pk_your_anon_key"

Filter operators

Operator	Example	SQL equivalent
eq	?status=eq.active	status = 'active'
neq	?role=neq.admin	role != 'admin'
gt / gte	?age=gt.18	age > 18
lt / lte	?price=lt.100	price < 100
like	?name=like.*john*	name LIKE '%john%'
ilike	?name=ilike.*john*	name ILIKE '%john%'
is	?deleted_at=is.null	deleted_at IS NULL
in	?id=in.(1,2,3)	id IN (1, 2, 3)

Insert rows

# Single row
curl -X POST http://localhost:8787/v1/db/users \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{"name": "Alice", "email": "alice@example.com"}'

# Multiple rows
curl -X POST http://localhost:8787/v1/db/users \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '[
    {"name": "Alice", "email": "alice@example.com"},
    {"name": "Bob", "email": "bob@example.com"}
  ]'

Update rows

Filters are required — you can’t update without a WHERE clause.

curl -X PATCH "http://localhost:8787/v1/db/users?id=eq.42" \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{"name": "Updated Name"}'

Delete rows

Filters are required — you can’t delete without a WHERE clause.

curl -X DELETE "http://localhost:8787/v1/db/users?id=eq.42" \
  -H "apikey: truss_sk_your_key"

Call functions

curl -X POST http://localhost:8787/v1/db/rpc/my_function \
  -H "apikey: truss_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{"arg1": "value1", "arg2": 42}'

Row-Level Security (RLS)

When using an anon key, Truss sets the Postgres role to authenticated and passes JWT claims via set_config. Your RLS policies can reference:

• current_setting('request.jwt.claims', true) — full JWT payload
• current_setting('request.jwt.sub', true) — the sub claim (user ID)

With a service_role key, RLS is bypassed entirely (role is set to postgres).

Dashboard features

The dashboard provides a full SQL workbench with:

• Monaco editor with syntax highlighting and auto-complete
• Schema browser — tables, columns, indexes, constraints
• ERD visualizer — auto-generated entity-relationship diagrams
• Saved queries — store and re-run frequently used queries
• Database branching — create isolated database branches for testing
• Backups — point-in-time recovery (PITR)