Docs menuCore Concepts
Core Concepts
SilentAuth is built around four primitives: Intents, Policies, Approval and Proof, and Receipts. Understanding how they interact is the key to integrating SilentAuth effectively.
The core flow: Your system declares an Intent → SilentAuth evaluates it against your Policies → If required, a human verifies and approves → A signed Receipt is issued → Your system validates the Receipt before executing.
Intents
An Intent is a declared request to perform a sensitive action. Before your AI agent, pipeline, or script can execute a high-risk operation, it declares what it wants to do, with which parameters, and which project or policy should evaluate it.
Policies
Policies are rules that determine which actions require human approval and under what conditions. You configure policies per-project and per-action. Policies can require approval from specific individuals, teams, or roles. They can also whitelist low-risk patterns (e.g., staging deploys) while requiring approval for production.
Approval and Proof
When an Intent matches a policy that requires human approval, SilentAuth sends an approval request to the designated reviewers. The reviewer sees the exact action and parameters, then approves or denies it from the dashboard. Stronger proof layers can be added for high-risk actions.
Receipts
A Receipt is the short-lived proof your agent or gateway verifies before execution. In Open Preview, receipts travel in the permit_token response field and include Receipt V2 claims such as params_hash, policy_hash, proof_tier, kid, expires_at, and revocation_status.