EXHIBIT 05● REAL HASH-DERIVED CHALLENGE
Fiat-Shamir (Non-Interactive Proof)
Turn interactive challenge-response into a single hash-bound proof transcript
✓ Completeness ✓ Soundness (ROM model) ✓ Practical NIZK pattern
Cryptographic honesty: challenge generation is real SHA-256 over transcript inputs and verification is real modular arithmetic. Parameters are intentionally small for readability.

In Schnorr, the verifier sends a random challenge c. In Fiat-Shamir, we remove that verifier message and derive c by hashing public transcript data. This turns an interactive proof into a single self-contained proof object.

That "hash as verifier" pattern is the bridge from educational Sigma protocols toward many modern signature and proof systems used in production.

Here the SHA-256 digest is reduced to only 50 challenges. Different messages can therefore produce the same challenge, and a tampered message can still verify. The result, log and narration report the computed equation and both challenges; passing this toy check does not establish secure message binding. RFC 8235 §2.3 describes cryptographic-sized hash challenges and contextual binding for Schnorr NIZK; this exhibit is a small pedagogical model, not a conforming production implementation.

Toy vs Production Parameters
DimensionThis ExhibitProduction Expectation
Hash inputR || y || messageDomain separated transcript with context binding
Challenge range1..50Large scalar field challenge
Math domainSmall prime modulusSecure elliptic curves or large finite fields
Security modelDidactic random-oracle intuitionAudited proofs and protocol-specific assumptions
Fiat-Shamir removes the live verifier: the challenge is derived by hashing the transcript itself, turning an interactive proof into a single non-interactive one.
PROVER OUTPUT
Message
—
Nonce commitment R = g^r mod p
—
Challenge c = H(R || y || message) mod 50 + 1
—
Response s = r + c·x mod (p-1)
—
VERIFIER CHECK
g^s mod p
—
R · y^c mod p
—
Ready.
— protocol log —
What changes from Schnorr: the verifier's random challenge is replaced by a deterministic challenge derived from hashing transcript data. This gives a non-interactive proof object at the cost of relying on random-oracle style assumptions.