Thread Vault CLI was built with privacy as a primary design constraint. This page documents exactly what data is stored, where, and when it is deleted.
All scraped data is stored in data/vault.db on your local machine. No data
leaves your device during scraping or unsending operations — only direct API calls to Instagram.
All scraped message rows, pagination cursors, session credentials, and thread cache are deleted automatically upon logout or Scrape Dashboard exit. This is a hard-coded behavior, not a setting.
Thread Vault CLI contains zero analytics, telemetry, tracking pixels, or error
reporting calls. The only outbound HTTP requests are to www.instagram.com — verifiable in
app/core/http.py.
Thread Vault never asks for, intercepts, or stores your Instagram password. Login
happens in a standard Chromium browser window pointed at instagram.com/accounts/login/ —
identical to logging in normally.
Thread Vault is MIT-licensed open source. Every line of code is available for review in the repository. You can verify exactly what data is stored, what network requests are made, and how the purge lifecycle works.
Session credentials are written atomically via temp-file swap in
app/core/cache.py. This prevents partial writes that could expose truncated credential data
if the process is killed mid-write.
Thread Vault makes
outbound requests exclusively to www.instagram.com. Here is the complete list of
endpoints used:
cache/session.json grants full Instagram access to whoever can read it.chmod 700 cache/. This
prevents other system users from reading your session file.app/core/http.py to audit network
calls, app/core/cache.py for credential storage, and app/core/db.py for the
purge lifecycle before trusting any CLI tool with your account session.