Shipyard

Use case · Safe team access

Database access for the whole team, without handing over prod.

People need to look at production data every day — debugging, support, analytics. Shipyard gives them a read-only path that can never mutate data, so access stops being a standoff between speed and safety.

The problem

Today, access is either blocked or dangerous.

Shared credentials everywhere

One prod password lives in a vault note half the team can open. Nobody knows who ran what, and rotating it breaks everything.

A stray UPDATE away from disaster

Engineers connect with clients that happily execute writes. One missing WHERE clause and you are restoring from backup.

Tribal SQL in DMs

The useful queries live in someone’s scratch file or an old Slack thread. Every investigation starts from scratch.

How Shipyard helps

Read access that is safe by construction.

Shipyard separates read from write at the connection level, so browsing and querying can never touch your data.

01

Read/write connection split

Each database gets separate read and write connections. Schema pulls and queries use the read connection — writes only happen via approved migrations.

02

SELECT-only Query Studio

Run read-only queries with table or vertical views. Engine-aware validation rejects anything write-shaped before it executes.

03

Saved and shared queries

Save queries with names, tags, and descriptions, and share them with the team via links — your tribal SQL, kept in one place.

04

Roles that match responsibility

Viewers browse and query, editors propose changes, admins manage access — every action tied to a real user, not a shared login.

How it works

From first connection to governed changes.

  1. Add a read connection

    Point Shipyard at a read-only credential for each database. Passwords are write-only once stored.

  2. Invite the team

    Give each person a role — viewer, editor, or admin — from the in-app Users console.

  3. Query safely

    The team explores schemas and runs SELECT-only queries in Query Studio across 22 engines.

  4. Build a shared library

    Useful queries get saved, tagged, and shared, so investigations start from knowledge instead of zero.

Open up your databases without opening up risk.

Read-only access, individual accountability, and a shared query library — in one tool.