Get Thread Vault CLI running in under 5 minutes. This guide covers the fastest path from zero to your first scrape session.
Thread Vault requires Python 3.13 or newer. Check with
python --version. Download from
python.org if needed.
The recommended way to install and run Thread Vault. Install with the one-liner below, or use pip with a manual venv.
Once the application launches, here's what happens and what you do:
Thread Vault silently checks for a cached session in
cache/session.json. If none exists or it's expired,
you're routed to the Auth Screen. If valid, you go straight to
the Thread Explorer.
Press A for Browser Login (Chromium
opens, log in normally, cookies are captured automatically) or
B to paste your
sessionid cookie value.
The Thread Explorer lists all your Instagram DM conversations. Use arrow keys to navigate, ctrl+f to search, r to reload from the API, then Enter to select.
The Thread Detail screen shows mode options. Press 1 (Media Archive), 2 (Text & Unclassified), or 3 (All Content). Confirm to enter the Scrape Dashboard.
In the Scrape Review Dashboard, press P to start the automatic pipeline — both the scraper and unsend worker run concurrently. Or use S (scrape only) and U (unsend only) for manual control.
Pressing Escape from the dashboard triggers the data isolation purge — all scraped rows, cursor state, and cache files are deleted from disk. Your session is clean.
data/vault.db when you exit the
dashboard or log out. If you need to retain data, export from the
SQLite database before exiting. See the
Export Guide
for instructions.
Detailed walkthrough of both Browser Login and Session ID authentication methods.
All dashboard features, manual vs automatic pipeline, rate limiting, and export options.
OS-specific setup guides for Windows, macOS, and Linux with troubleshooting sections.