What does this research mean for the field?

The SZL Doctrine v2 formalizes a verbal doctrine into a machine-checkable, self-enforcing contract that binds autonomous agents to strict quality and artifact standards. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The aim is to formalize a verbal doctrine into a machine-checkable contract for SZL agents.

May 16, 2026Open Access

SZL Doctrine v2 — 9 Canonical Axes (Λ DOI)

Key Points

The aim is to formalize a verbal doctrine into a machine-checkable contract for SZL agents.
Defined 9 quality axes and 6 artifact dimensions with a scoring rubric.
Implemented self-enforcement mechanisms where agents include a grading block.
Configured to reject outputs that do not meet the minimum score requirement.
Agents must achieve scores ≥ 0.9 on defined axes to generate accepted outputs.
Self-grading ensures accountability, improving output quality.
Failure to comply with the §4 Self-Grade block results in automatic rejection.

Abstract

Doctrine v2 formalises the original six-clause verbal doctrine of Stephen P. Lutar Jr. into a machine-checkable contract binding every SZL agent (Alloy, a11oy code-agents, fabric agents, sentra, amaru). The spec defines 9 quality axes (cleanliness, horizon, resonance, frustum, gaussClosure, invariance, moralGrounding, ontologicalGrounding, measurabilityHonesty) and 6 Λ₁₀ artifact dimensions (CODE, CODEX, API, TEST, THESIS, SURFACE), each with a 0-1 rubric, measurable proxy, and Lean theorem stub. Every axis and dimension requires a score ≥ 0.9 before output is accepted. The doctrine self-enforces: agents are injected with the contract at spawn time and must include a §4 Self-Grade block in every output; PRs without this block fail CI automatically.

SZL Doctrine v2 — 9 Canonical Axes (Λ DOI)

Key Points

Abstract

Cite This Study