> ## Documentation Index
> Fetch the complete documentation index at: https://docs.webrayn.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Session Strategy

> How to preserve authentication and browser state across scrapes — reuse a persistent profile, keep the session alive, re-attach authenticated Chrome, and transfer cookies between engines.

# Session Strategy

The question: *how do I keep a login and browser state alive across many pages
and many tool calls?*

## Preserve, don't recreate

* Use ONE persistent profile per account (`webrain_session(op=login)`).
* Keep the session open across pages (`webrain_session(op=open)` → navigate →
  navigate → … → close).
* Never discard a working profile/session — that is how you end up re-solving
  challenges.

## Authenticate once

1. `webrain launch <service> <profile> <url>` (or `webrain_session(op=login)`).
2. Log in (native login, vault + TOTP) or let the human complete a 2FA/approval
   gate in the headed browser.
3. The profile now carries the session.

## Reuse it

* Re-attach the authenticated Chrome: `CDP_URL=http://127.0.0.1:9222` (or
  `webrain_session(op=open, cdp_url=...)`).
* Every subsequent navigate/extract shares the session.

## Transfer to another engine

Export cookies on the live browser (`webrain_session(op=cookies)`), import on the
target connection (`op=setcookies`), then batch on that same connection. Or use
`op=save_state` / `op=restore_state` for portable cookies + localStorage.

## Session isolation

Different sessions/browsers have isolated cookie jars. Set cookies and batch on
the SAME connection — obscura keeps per-connection cookie jars, so a fresh
connection starts empty.