Deterministic Knowledge Infrastructure for AI Systems.
DigiEmu Core provides a deterministic snapshot and verification layer for knowledge states.
It allows AI and software systems to store, replay, verify and sign knowledge states with full reproducibility.
The system ensures that knowledge snapshots can always be reconstructed and verified.
- Deterministic knowledge snapshots
- Cryptographic integrity
- Replay verification
- Signed knowledge bundles
- Artifact origin metadata
- Transportable bundles
DigiEmu Core verifies deterministic decision-state integrity through canonical JSON, SHA-256 snapshot hashing, and replay/verify checks. It does not certify agent identity, trustworthiness, authorization, action legitimacy, or legal compliance. These concerns belong to complementary external trust and attestation layers such as TBN.
DigiEmu Core is moving toward a public standard structure for deterministic AI decision verification.
- DigiEmu Core Specification v0.9
- Test Vectors v0.9
- Negative Test Vectors v0.9
- Test Vector Manifest v0.9
- Conformance v0.9
- Conformance Declaration v0.9
- Conformance Declaration Schema v0.9
- Verify Report Examples v0.9
- Spec Index v1.0
- CLI Contract v1.0
- Snapshot Hash v1.0
- Verify Spec v1.0
- Verify Result Schema v1
- Verify Report Schema v0.9
- Snapshot Bundle v1.0
Specification -> Test Vectors -> Negative Test Vectors -> Test Vector Manifest -> Conformance -> Conformance Declaration -> Conformance Declaration Schema -> Verify Report Examples -> Verify Report Schema
The specification explains the model.
The test vectors make verification reproducible.
The conformance document defines implementation requirements.
The verify report examples define machine-readable verification outcomes.
go build -o digiemu ./cmd/digiemudigiemu snapshot file input.jsondigiemu verify bundle snapshots/.../bundle.jsondigiemu replay bundle snapshots/.../bundle.jsondigiemu verify replay snapshots/.../bundle.jsondigiemu sign bundle bundle.jsondigiemu verify signature bundle.jsondigiemu identity showdigiemu identity export <directory>digiemu identity import <directory>digiemu identity fingerprintdigiemu export bundle bundle.json <directory>digiemu import bundle <directory>DigiEmu Core guarantees that a knowledge snapshot can always be:
- recreated
- replayed
- verified
- cryptographically signed
- transported
- trusted via identity verification
DigiEmu Core consists of:
- Snapshot engine
- Replay engine
- Verification layer
- Signature system
- Artifact origin / signer metadata utilities
- Bundle transport system
cmd/digiemu
pkg/snapshot
pkg/replay
pkg/verify
pkg/claims
pkg/meaning
pkg/uncertainty
DigiEmu Core v1.0.0 is the first deterministic baseline release.
The system supports:
- deterministic snapshot creation
- replay verification
- cryptographic integrity validation
- origin-bound or externally attested artifacts
Enterprise hardening and ecosystem integration are ongoing.
Current version: v1.0.0
See:
https://github.com/DigiEmu/core/releases
Business Source License 1.1.
See LICENSE for details.
Status: draft / pre-release / not stable
Core 2.0 Draft 1 (tag example: v2.0.0-draft.1) is a partner-testable milestone.
It is intended for evaluation and feedback; it is NOT a stable release and does
not change any v1.0 behavior.
Short positioning
- DigiEmu Core provides deterministic knowledge infrastructure for AI systems.
- Core 2.0 Draft 1 is partner-testable and draft-only.
- v1.0 behavior remains unchanged and is the baseline for compatibility.
Core 2.0 capabilities (draft)
- Experimental conformance CLI
- Human-readable conformance output
- JSON conformance reports
- Conformance report schema
- OpenAPI contract draft
- Docker usage path (documentation)
- CI conformance checks
Quickstart commands (for evaluation)
go test ./...
go run ./cmd/digiemu experimental conformance run testdata/core_2_conformance
“DigiEmu Core by Bruno Baumgartner © 2026”
go run ./cmd/digiemu experimental conformance run testdata/core_2_conformance --jsonDocker (optional evaluation)
docker build -t digiemu-core .
docker run --rm digiemu-core experimental conformance run /opt/testdata/core_2_conformance
docker run --rm digiemu-core experimental conformance run /opt/testdata/core_2_conformance --jsonExpected output
Conformance run summary: total=11 passed=11 failed=0
Important notes
- Core 2.0 remains draft unless explicitly marked stable.
- Experimental commands remain under the
experimentalnamespace. - This is not a stable
v2.0.0release.
Links
- Conformance Quickstart
- Partner Integration Notes
- Docker Usage
- OpenAPI Draft
- Release Checklist
- Tagging Plan
“DigiEmu Core by Bruno Baumgartner © 2026”