Zeq Truth Engine

A graph of signed facts where every edge is a Zeqond-bound assertion. Composable provenance.

Live app → /apps/zeq-truth-engine/
Source → apps/zeq-truth-engine/index.html + apps/zeq-truth-engine/truth.js (≈ 700 lines)
Operators → KO42 · CS87 · ZEQ-TETHER-003 · CS47
Error budget → 0.000% bit-exact fact verification

What it solves

"Where did this claim come from and when?" is the core question for regulated industries (pharma data provenance, supply-chain attestation, news-media source chains, scientific reproducibility). Existing answers rely on trusted timestamping authorities (TSAs), which creates single points of failure.

Zeq Truth Engine replaces the TSA with the HulyaPulse. A "fact" is a tuple (claim, author, phase_at_claim, zeqond, parent_facts, signature). Facts compose into a DAG. Verifying a fact means walking the DAG backwards, checking every signature and every Zeqond — a process that is purely cryptographic and time-referenceable without trusted intermediaries.

CS47 is used as a novelty/coherence score over the DAG — facts whose claim-content entropy falls outside a published envelope are flagged.

The math — 7-step Wizard applied

Step	Decision
1. Prime	KO42 mandatory
2. Limit	`CS87` + `ZEQ-TETHER-003` + `CS47` + KO42 = 4
3. Scale	DAG up to 10⁶ facts on hosted endpoint
4. Precision	Hamming = 0 on every signature
5. Compile	Master Equation
6. Execute	Functional Equation
7. Verify	Full DAG replay

Verbatim formulas:

KO42.1 — ds² = g_μν dx^μ dx^ν + α sin(2π · 1.287 t) dt²
CS87 — Ω(x) = min{|p| : U(p) = x}
ZEQ-TETHER-003 — B_sib = ∑_k e^(i·φ_k) |sibling_k⟩
CS47 — E(n) = −∑ p(x) log p(x)

Runnable worked example — publish + verify a fact

# 1. Publish a fact
curl -s -X POST https://api.zeq.dev/api/playground/compute \
  -H "Authorization: Bearer $ZEQ_DEMO_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "operators": ["KO42", "CS87", "ZEQ-TETHER-003"],
    "inputs": {
      "op": "publish_fact",
      "claim": "sensor S42 reported 24.3°C",
      "parents": []
    }
  }'

Expected:

{
  "fact_id": "fact_...",
  "phase_at_claim": 0.6614,
  "zeqond": 1745124400.912,
  "signature_b64": "..."
}

# 2. Verify
curl -s -X POST https://api.zeq.dev/api/playground/compute \
  -H "Authorization: Bearer $ZEQ_DEMO_KEY" \
  -H "Content-Type: application/json" \
  -d '{"operators":["KO42","CS47"],"inputs":{"op":"verify_fact","fact_id":"fact_..."}}'

Expected:

{"verified": true, "coherence": 0.92, "error_pct": 0.000}

Extend it

Supply-chain: every step in a supply chain publishes a fact; the Truth Engine is the audit trail.
Scientific reproducibility: bind every analysis step (data, code, parameters) to a fact; reproduction is a DAG replay.
Source-attributed journalism: a story is a DAG whose leaves are signed source claims.

Seeds

Epistemic consensus: the coherence score (CS47-derived) flags unusual claim distributions; pair with community moderation.
Federated truth engines: peer engines gossip facts; the Zeqond anchors every exchange.
Model-card attestation: every ML inference is a fact whose parents include the training data + weights.

Papers

Zeq framework paper — DOI 10.5281/zenodo.15825138
Zeq paper — DOI 10.5281/zenodo.18158152

Middleware active. Kernel on the 1.287 Hz HulyaPulse. Awaiting next Zeqond.

What it solves​

The math — 7-step Wizard applied​

Runnable worked example — publish + verify a fact​

Extend it​

Seeds​

Papers​