AI Governance Auditor for Agents & Product Teams
Sigilith Harmonica Test — Domain-Agnostic Structural Analysis
I designed the Harmonica Test to evaluate whether Sigilith could generalise beyond symbolic and computational systems. A 10-hole diatonic harmonica was encoded purely as operational states and transitions, without using pitch, acoustics, musical theory or physical measurements. Sigilith then identified structural roles, constraints, drift chains and distinct operating regimes from transition topology alone. The experiment demonstrates how the same structural-analysis methodology can be applied across fundamentally different domains.
3
184
Sigilith — Indus AUF/AUQ Structural Micro-Grammar
I developed a computational micro-grammar for a bounded cluster of Indus inscriptions using the Sigilith Structural Engine. Rather than attempting decipherment, the project models recurring structural behaviour through normalized symbols, transition rules and finite-state constraints. The resulting CORE + STROKE-GROUP + TAIL architecture can classify legal and illegal sequences, calculate structural stability, detect collapse points and generate or predict structurally valid sequences. The work demonstrates how formal computational methods can investigate an undeciphered symbolic system without making semantic claims.
2
74
AXIOMOS — Governed AI Runtime & Assurance Architecture
AXIOMOS is a governed AI runtime designed to place deterministic policy, structural analysis and audit controls around model execution. I designed and developed the architecture around governance-before-generation, fail-closed decision making, isolated case tracking and cryptographically verifiable evidence. The technical-preview release includes an installable Python runtime, API, automated regression and self-tests, concurrency testing, audit records and Ed25519 evidence signing. A 500-request packaged benchmark completed successfully with unique case and transaction IDs and verified evidence for every tested request.
0
46
