Crate sigil_protocol 

SIGIL — Sovereign Identity-Gated Interaction Layer

An open protocol for securing AI agent-to-tool interactions.

SIGIL defines traits (interfaces) for:

• Identity — binding users to trust levels
• Scanning — detecting sensitive content before it enters agent context
• Vault — encrypted storage for intercepted secrets
• Audit — tamper-evident logging of all security events
• Policy — permission and rate-limiting enforcement

Implement these traits with your own backends (regex, HSM, LDAP, etc.) to add SIGIL-compliant security to any agent system.

Quick Start

use sigil_protocol::{SensitivityScanner, AuditLogger, IdentityProvider, SecurityPolicy};
// Implement these traits with your own backends

Re-exports

pub use audit::AuditEvent;

pub use audit::AuditEventType;

pub use audit::AuditLogger;

pub use identity::IdentityBinding;

pub use identity::IdentityProvider;

pub use identity::TrustLevel;

pub use policy::RiskLevel;

pub use policy::SecurityPolicy;

pub use scanner::SensitivityScanner;

pub use sigil_envelope::SigilEnvelope;

pub use sigil_envelope::SigilKeypair;

pub use sigil_envelope::Verdict;

pub use vault::VaultEntry;

pub use vault::VaultProvider;

Modules

audit

Audit — tamper-evident logging of all security events.

identity

Identity — binding users to trust levels.

mcp_server

Reference SIGIL MCP Server.

policy

Policy — permission and rate-limiting enforcement.

registry

Remote pattern fetching from the SIGIL community registry.

scanner

Sensitivity scanning — detect sensitive content before it enters agent context.

sigil_envelope

SIGIL Envelope — per-message DID-bound cryptographic signing.

vault

Vault — encrypted storage for intercepted secrets.