A local web app for ThoughtSpot admins

Clean up ThoughtSpot
without writing a script.

Find stale content, clean up leavers, and see what depends on what. The toolkit caches your instance's users, groups and content so you can browse instantly, then runs bulk changes with a dry-run preview and a TML backup first.

Install in one command View on GitHub

For ThoughtSpot Cloud or Software. No Python or Node.js needed.

The Dashboard: what needs attention, user, group and content totals, cache freshness per entity, content by type and recent jobs

1 commandto install; it brings its own Python

3 platformsmacOS, Linux and Windows

3 sign-insBasic, Trusted Auth or Bearer token

127.0.0.1runs on your machine only, never exposed

Get started in three steps.

You need a ThoughtSpot login with the ADMINISTRATION privilege. Everything after the install happens in the browser.

Install

curl -LsSf https://raw.githubusercontent.com/BibekTS/thoughtspot-admin-toolkit/main/install.sh | sh

powershell -ExecutionPolicy Bypass -c "irm https://raw.githubusercontent.com/BibekTS/thoughtspot-admin-toolkit/main/install.ps1 | iex"

Installs into its own isolated environment. Re-run it to upgrade. Behind a corporate proxy? See the certificate fix.

Open the app

ts-admin-toolkit serve

Opens http://localhost:8080 in your browser. Keep the terminal open while you use it.

Connect your instance

Enter your ThoughtSpot URL, pick a sign-in method and click Test connection.

BasicTrusted AuthBearer token

Credentials go to your OS keychain. Then sync each section on its own; nothing forces a full sync.

What you can do.

Browse from a local cache, then act in bulk. Every sync and bulk operation runs as a tracked background job with live progress.

Archiver: liveboards and answers unused for 90 days, with owner, tags, days unused and views — **Archiver**Find stale liveboards and answers, tag them, delete with a TML backup, and restore from History.

Users: a searchable grid of users with status, creator and created date — **Users**Transfer ownership, transfer sharing and bulk delete leavers. A drawer shows each user's groups and effective privileges.

Metadata: all content with type, connection, owner, tags, last accessed and views — **Metadata**Search and filter all content by owner, tag, last accessed and views, and export it to CSV.

Lineage: a connection feeding logical tables, a model, and the liveboards and answers built on it — **Lineage**A dependency graph, plus column-level lineage from a model column back to its physical table and forward to every consumer.

Bulk DeleteDelete by GUID or search, with dependency resolution.

Bulk SharingShare content to users and groups in bulk, with an impact preview.

GroupsBrowse groups, privileges, members, and who created each group. Read-only.

JobsEvery sync and bulk operation is a tracked background job with live progress.

Multi-instanceSeveral ThoughtSpot instances, each with its own cache, one picker in the sidebar.

DiagnosticsTail logs and download a support bundle with no credentials in it.

Screenshots use sample data.

How it works.

The toolkit talks to the public ThoughtSpot REST API v2 and mirrors users, groups, metadata and lineage into a local SQLite database. Browsing reads the cache. Writes go straight to ThoughtSpot.

Dry run firstEvery destructive action previews its full impact before you confirm.

TML backupDeletions are backed up to ~/.ts-admin/tml-exports/ and can be restored.

Live checkWrites are re-verified against ThoughtSpot at execution time, not against the cache.

Audit logEvery write is recorded.

Install it in one command.

Then run ts-admin-toolkit serve and connect your instance in the browser. Try it on a non-production instance first.

Get started Releases

Clean up ThoughtSpotwithout writing a script.