Loading…
My personal AI activity
Control what an agent may buy, access or execute on your behalf.
The same protocol, pointed at decisions you make for yourself rather than ones an enterprise makes at scale. Shield governs disputed charges and protection limits today; Sense — a browser-side view of what your agents are doing — is in development.
What goes wrong
The specific actions teams tell us they are afraid of. Each one is a decision that happens faster than a human can review it.
- An agent buying something you did not approve
- An assistant reaching an account it should not
- A subscription renewing past a limit you set
- A disputed charge you need evidence for
Where the decision goes
One deterministic check between the intent and the effect — ALLOW, ESCALATE or BLOCK, with a signed record either way.
An intent — from an AI, a user, or a system that wants to act — reaches Decision, where the Decionis Protocol evaluates the action. The evaluation returns one of three verdicts: ALLOW, meaning authorized; ESCALATE, meaning it needs approval or more information and is routed to whoever holds the authority; or BLOCK, meaning it is high risk or out of policy. An allowed action reaches Execution, where it is executed safely and recorded; an escalated action reaches it only once approved; a blocked action never does. Decionis is the neutral decision authority between intent and execution.
Surfaces — where you encounter it
Connectors — how systems reach it
Shield is live. Sense is in development and not yet available — nothing on this page depends on it.
What to do next
The same five steps every time. Nothing else on this page competes for your attention.
- Simulate this recipe
Run the policy against a realistic action in the browser. Push it past what the rules allow and watch the verdict come back. No account.
- View the signed Decision Dossier
See exactly what was decided and why: the rule that fired, the evidence it read, the policy version in force, and an Ed25519 signature you can verify yourself.
- Deploy in shadow mode
Measure what the policy would have caught on your own traffic without touching the live path. Every recipe defaults to shadow, so the first deployment carries no execution risk.
- Connect a real workflow
Point the same policy at the system where the action actually originates — a checkout, an ERP posting, a Zap, an agent's tool call.
- Share the verified result
Publish the proof: a public verification link, an embeddable badge, a PR comment, or an anonymized shadow-mode finding. This is how the next person discovers Decionis.