How an AI agent uses Encrypted identity vault
Per-site credentials, encrypted on your device, never written in the clear.
Updated · T3rnel Browser 1.3.0
What the agent can do
- AES-256-GCM with a key derived from your master passphrase.
- A wrong passphrase fails the authentication tag and is refused, rather than returning plausible garbage.
- Every vault operation requires confirmation: a write can overwrite a live credential and a delete destroys the only copy, so neither is treated as safer than a read.
The tools
| Tool | Risk | Tier |
|---|---|---|
browser_vault_save | high | Pro |
browser_vault_load | high | Pro |
browser_vault_list | low | Pro |
browser_vault_delete | high | Pro |
Calling it over MCP
{ "method": "tools/call", "params": { "name": "browser_vault_save", "arguments": {} } }
Connect a client first: add the MCP Session Bridge. The extension then exposes 97 tools against your signed-in tabs.
Questions
Which tools does Encrypted identity vault use?
browser_vault_save, browser_vault_load, browser_vault_list, browser_vault_delete.
Is Encrypted identity vault free?
Partly. browser_vault_save, browser_vault_load, browser_vault_list, browser_vault_delete need Pro ($29.99 once); none of the others work on the free tier.
Does the agent have to ask before using Encrypted identity vault?
Each tool has a risk level: browser_vault_save is high; browser_vault_load is high; browser_vault_list is low; browser_vault_delete is high. The approval gate is off by default and can be armed in Settings.