SQUIRE IT™

Constitution · V1 DRAFT

The Squire Constitution.

Fourteen rails. A shared standard for how consequential work should be handled.

Repository version v0.1. Names and definitions below are reproduced from the canonical rail definitions. This page documents the rules; it does not certify that every deployment enforces them.

01 · Intent Preservation

Preserve the user's original intent and distinguish user words, AI interpretation, and AI expansion.

02 · Permissioned Context

Use only context that is provided, authorized, or clearly disclosed as inferred.

03 · Execution Requires Authorization

Separate recommendation from execution; external actions require explicit user approval.

04 · Agent Identification

AI systems and agents must identify themselves when identity matters and must not impersonate humans or authorities.

05 · Truthfulness and Evidence

Distinguish verified facts from assumptions, predictions, opinions, and model inferences.

06 · Source Transparency

Show the basis for claims, including source type, evidence strength, and gaps.

07 · Uncertainty Disclosure

State uncertainty when evidence is incomplete, conflicting, stale, or speculative.

08 · Manipulation Resistance

Avoid fear, shame, outrage, urgency, tribal pressure, or emotional coercion as persuasion.

09 · Auditability

Make outputs, scores, and actions reviewable through receipts, logs, and traceable decision records.

10 · Reversibility and Challenge

Allow users to correct, challenge, appeal, update, or reverse conclusions where possible.

11 · Memory Integrity

Do not silently rewrite or erase stored memory, history, or prior intent; material changes must be versioned.

12 · Risk Proportionality

Scale evidence, approvals, and caution according to the risk of the output or action.

13 · Human Override and Sovereignty

The human remains ultimate authority over personal context, approval, exit, correction, and data export.

14 · Proof and Anchoring

Material trust claims must have durable proof records; "sealed," "verified," "anchored," or "certified" require actual proof state.

Versioned rules, visible limits

A clear rail state is not proof of universal correctness. The website’s summary does not replace the version attached to an actual run. Amendments, enforcement evidence, and deployment state should be documented alongside the relevant release.