Open-Source Protocol
IDProva
Verifiable identity for the agent era
IDProva is an open protocol for establishing cryptographically verifiable identity, scoped delegation, and tamper-evident audit trails for autonomous AI agents. Built by Tech Blaze Consulting and submitted to NIST CAISI.
Three Pillars
Every AI agent interaction needs answers to three questions. IDProva provides all three.
Identity
Who is this agent? W3C DID-based Agent Identity Documents with cryptographic keys, agent metadata, and trust level declarations.
Delegation
What can it do? Delegation Attestation Tokens — signed, scoped, time-bounded, chainable permission tokens with constraint inheritance.
Audit
What did it do? Hash-chained Action Receipts providing tamper-evident audit trails mapped to NIST 800-53, ISM, and SOC 2 controls.
Why IDProva
Post-Quantum Ready
Hybrid Ed25519 + ML-DSA-65 (FIPS 204) cryptography from day one.
Compliance-Mapped
Action Receipts satisfy NIST 800-53, Australian ISM, and SOC 2 controls out of the box.
Open Protocol
Apache 2.0 licensed. No vendor lock-in. Self-host or use managed registry.
NIST-Aligned
Submitted to NIST CAISI (NIST-2025-0035). Participating in NCCoE AI Agent Identity project.
Protocol Bindings
First-class integrations with MCP, A2A, and HTTP transports.
Written in Rust
Memory-safe, high-performance implementation with SDKs for Python, TypeScript, and Rust.
Built by Tech Blaze Consulting
IDProva brings deep government security and IRAP assessment expertise to the AI agent identity challenge. The same rigour we apply to ISM compliance, now applied to autonomous agent systems.