Evaluation defines the heart of Attestor: how Attestations and eligible governed inputs undergo Rule-Constrained Evaluation to produce a bounded Trust Statement.
Attestor does not convert sources directly into conclusions. It evaluates a defined assertion against an eligible, attributable, provenance-preserving basis under governed rules and states only the conclusion that basis can responsibly support.
Rule-Constrained Evaluation is the governed process by which Attestor examines an Attestation against its eligible evaluation basis and determines what bounded conclusion Attestor can responsibly state. Evaluation is the institutional bridge between assertion and Trust Statement.
Every evaluation must identify the Attestation and assertion being evaluated. Attestor evaluates a defined assertion within a defined scope; it does not perform an unbounded search for whether a subject is generally trustworthy or true.
The evaluation basis consists only of eligible governed inputs admitted for the particular evaluation. Material supporting, limiting, conflicting, contradictory, and indeterminate information must remain available to the evaluation rather than being filtered according to the desired result.
Evaluation must operate under identifiable Attestor rules and methodology. Rules constrain what may be considered, how scope and authority are preserved, how uncertainty is handled, and what conclusion the available basis permits.
Attestor must preserve enough reasoning context to make the treatment of material inputs reviewable. Evaluation is not a mechanical vote among sources, an authority popularity contest, or a hidden confidence calculation.
The evaluation conclusion must be no broader or stronger than the eligible basis, applicable scope, provenance, authority, relevant state, limitations, and uncertainty support. Where the basis cannot justify a stronger conclusion, Attestor must preserve that limitation.
The controlled outcome classifies the evaluation result as supported, partially-supported, not-supported, contradicted, or indeterminate. The outcome summarizes the bounded evaluation; it does not replace the Trust Statement conclusion.
A completed evaluation provides the governed basis for a Trust Statement. The Trust Statement remains a distinct Attestor-owned canonical object with its own identity, scope, attribution, provenance, limitations, status, and relationships.
The evaluation process is governed so that Attestor cannot reach a conclusion before the assertion, scope, eligible basis, authority and provenance context, and applicable rules are known. Methodology now defines the standing operational sequence, while Validation supplies machine-testable controls for the governed objects and context used in production.
The first controlled production operation exercised Rule-Constrained Evaluation against ATT-2026-0001 using a separately assembled Evaluation Basis composed of eligible governed inputs. The evaluation treated the asserted identity, origin, relevant certification state, and Suite relationships as bounded components rather than collapsing them into a generalized trust judgment.
The production result demonstrates the evaluation method; it does not establish that future Attestations will receive the same outcome. The supported result belonged only to this bounded assertion, evidence basis, scope, authority context, relevant state, and applicable rules.
The Controlled Values architecture establishes five evaluation outcomes. Their semantics govern production Evaluation. Methodology defines how the evaluation is conducted, while Validation tests applicable structural and normative requirements without replacing the Evaluation itself.
The controlled outcome is a classification of the evaluation result. The Trust Statement conclusion is the bounded human- and machine-interpretable statement Attestor actually makes. Two evaluations may share an outcome class while requiring different conclusions because their subjects, scopes, evidence, limitations, or relevant states differ.
No source object, certification, registry record, historical entry, integrity reference, discovery signal, intelligence product, workflow result, or external assertion automatically becomes an Attestor conclusion. Each remains within its source authority and may enter evaluation only when eligible.
This architecture does not adopt a universal numeric weighting system, confidence percentage, reputation score, majority-source rule, or automatic authority hierarchy. If future specialized methodology requires weighting for a defined evaluation type, that weighting must be explicitly governed, explainable, bounded to that methodology, and separately adopted.
Material disagreement, incomplete evidence, unresolved provenance, temporal mismatch, source-state change, and other uncertainty must influence the conclusion when relevant. Attestor does not force every evaluation into a positive or negative answer. indeterminate is a legitimate governed outcome when the basis cannot responsibly support a stronger determination.
The evaluation must remain tied to the relevant state and time of its inputs. A later change to a source object does not retroactively rewrite what Attestor evaluated. Material later changes may instead trigger review, correction, supersession, or a new evaluation under Lifecycle and Versioning rules.
A governed evaluation must preserve enough context for a later reviewer to understand what was evaluated, what basis was admitted, what material information was excluded and why when relevant, what rules applied, what limitations existed, and how the resulting conclusion remained within those bounds.
The first production evaluation concluded that the eligible governed evidence supports the asserted canonical identity and attributable Satoshium Certifier origin of SC-CERT-2026-0001, its representation by the canonical Certifier source as Issued · Active within the evaluated state and scope, and its traceable governed relationships to SREG-2026-0001, CHR-2026-0001, ANCH-2026-0001, and BEAC-2026-0001.
The conclusion expressly preserved the authority of each referenced Suite institution and did not establish the substantive truth of the underlying Atlas jurisdiction intelligence, independently re-certify the Certifier decision, extend Anchor integrity beyond its defined SCRD representation, establish unchanged state beyond the evidence reviewed, or create a generalized determination of trustworthiness.
Attestor Evaluation does not declare universal truth, inherit another institution's authority, certify on behalf of Certifier, create generic reputation, produce a trust score, guarantee future conditions, or erase uncertainty. Its responsibility is narrower and more defensible: produce an attributable, reviewable, bounded Attestor conclusion from an eligible governed basis.
Evaluation is Attestor's central institutional act and is now established both architecturally and operationally. The first controlled production operation demonstrated that an Attestation can be evaluated against a separately governed eligible basis under applicable rules to produce a controlled outcome and bounded Trust Statement conclusion without converting source authority into Attestor authority.
Production proof establishes that the evaluation architecture can govern a real Attestor determination. It does not predetermine any future Evaluation Outcome, evidence treatment, conclusion, Validation result, Conformance determination, lifecycle state, or Publication decision.
Back to Attestor →