It outlives the chat
A key in a prompt is in a log, a sync, and someone else's retention window.
One local vault. Agents in Cursor, Claude Code, Codex, and VS Code ask for a capability. A Meta model (Muse Spark by default) scores the request. You approve when policy requires it. The secret never enters the chat.
A key in a prompt is in a log, a sync, and someone else's retention window.
Rotation starts after the leak. The bill does not wait for your dashboard.
Agents ask all day. A password manager prompt was not designed for that.
AES-256-GCM. Master key in the vault you own. Hard ceilings. Secure even with no model at all.
Default Muse Spark 1.3 via the Meta API, or a model on your own local endpoint. It sees the request, never the key.
After the call, after 60 seconds, or on one revoke. Standing approval is a low read only.
Explore the request flow below. On phones, swipe the diagram sideways.
Nothing is stored. These buttons only show which gate fires.
Agent, IDE, alias, scope, reason. That is the whole prompt.
The policy floor. A write under a read ceiling is dead before the model runs.
Twenty-seven surfaces. The build fails if the secret shows up. The sweep.
The agent can ask, list aliases, enroll, check status, and revoke. The snippet.
From the crypto industry, securing wallets since 2025.
You know where the keys at. Always.