How to use Crudy

This guide walks you through Crudy from setting up a connection to editing records, building views and reviewing the audit trail. If you have not installed the app yet, see how to install Crudy first.

1. Add a connection

Open Crudy settings → Connections and click Add connection. Give it a clear title (for example "ERP production database"), choose who can access it, and save. Crudy generates a unique bearer token for this connection and reveals the agent setup instructions.

2. Create the database user

On your database server, run the SQL shown in the connection modal to create a least-privilege user. Crudy only needs four privileges — SELECT, INSERT, UPDATE and DELETE — on the single database you want to browse. Replace the placeholder password and database name with your own.

3. Install the agent

Download the agent binary for your server's architecture (use uname -m if you are unsure which one), copy it to the database server, and run the install command. The command already includes your connection's token; just fill in the database host, user, password and name. Run it as root and the agent installs itself as a service that starts on boot and reconnects automatically. When it connects, the connection shows as online.

4. Browse tables

Open Crudy from the navigation and choose your connection. The left sidebar lists every table in the database. You can:

• Search the table list — type to filter, and regular expressions are supported for precise matches.
• Pin the tables you use most so they stay at the top in a dedicated Pinned section.
• Click a table to load its records on the right.

5. Browse and search records

Records load page by page and stay fast even on very large tables. Every column is sortable, and the search box searches across all fields, not just the primary key. Use Refresh to pick up rows added since you opened the table (including rows added by other applications).

6. Edit, create and delete records

• Edit — double-click any row to open it in a form, change the fields, and save.
• Create — click Add record, fill in the fields (auto-increment keys are handled for you), and save. New rows appear immediately.
• Delete — open a record and confirm the delete. The row is moved to the Crudy trash, not destroyed.

Every change is validated and written to the audit log.

7. Build saved views

Need a focused slice of a table? Click New view to open the view builder. Choose the columns you want (use Add all fields to add them all at once), apply optional aggregates, add filter conditions, group by, order by, and set an optional row limit for top-N results. Save it and the view becomes its own tab you can re-run any time. Views are private to each user, and editable views let you double-click a result row to open the underlying record.

8. Save record filters

On the records view you can build multi-condition filters (field, operator and value) and save them by name. Saved filters are stored per user on the server, so they follow you to any device. Apply or clear them at any time.

9. Recover from trash

Open Crudy trash to see deleted rows with a preview, who deleted them and when. Restore a row back into its table either with its original ID or a fresh auto-increment ID. If the original ID has been reused, Crudy reports the conflict instead of overwriting.

10. Review the audit and error logs

The Audit log records every insert, update, delete and restore with the user, the table, the record ID and the old and new values; open any entry for the full detail. The Error log captures the raw database error and the statement whenever an operation is rejected, so you can diagnose problems quickly.

11. Export (admins)

Administrators viewing records get Excel export and print buttons for the records on screen, useful for quick reports and sharing.

Tips

• Pin your everyday tables so you never scroll the full list.
• Use saved views for the questions you ask often instead of re-building filters each time.
• If a connection shows offline, check that the agent service is running on the database server.