Skip to content

What must be true before this promise can trigger an action?

Assurance Linker turns one structured machine promise into a checklist for evidence, authority, action coverage, capital, recourse, and the named consequence.

The promise and the relying party’s verification context remain separate. Bulla reports whether the supplied records meet the supplied requirements; it does not decide whether the underlying claim is true.

Technical status for the source-only profile.

Current status: experimental, SOURCE_ONLY, synthetic-public, and team-operated.

What this check cannot tell you

The profile and Trial 0.2-r5 instrument use team-authored synthetic promises, cases, expected projections, implementations, evidence, keys, witness roots, rail observations, and settlement roles. No two-mirror publication set, cold-reader calibration, or foreign checker attempt has occurred, and recruitment readiness remains blocked. Surface completeness is relative to named frozen schemas, AL-* rules, and protected fields; it does not cover every open-world interaction. Foreign authorability, independent reproduction, institutional handling, operational separation, and marginal decision value remain blocked. Recruitment is an investigator-selected, uncompensated convenience sample and does not establish economic adoptability or representative demand. Source-to-image correspondence remains NOT_COMPUTED without a reproducible rebuild. The evidence does not establish worldly truth, complete denominators, external collateral encumbrance, custody, collectibility, legal enforceability, organizational independence, production safety, actuarial calibration, or mainnet readiness. Bitcoin is an optional rail adapter and does not evaluate predicates. External A/J/I/W and replay counts remain 0/0/0/0 and r0.

CHECKED FIXTURE COMPARISON

One promise. One action without a receipt.

Release 125,000 sat when the fixed artifact is delivered within bounds, required receiver-effect coverage is complete, authority is accepted, and the 200,000 sat allocation is adequate.

Complete record

Receipt integrity
VERIFIED
Receiver record
effect-mediated-001
Receipt set
effect-mediated-001
Coverage
COVERED
Capital
ALLOCATION_ADEQUATE
Payment
ELIGIBLE

Direct bypass

Receipt integrity
VERIFIED
Receiver record
effect-mediated-001, effect-bypass-001
Receipt set
effect-mediated-001
Coverage
UNCOVERED
Capital
ALLOCATION_ADEQUATE
Payment
INELIGIBLE under the supplied policy and verification context

Complete record

COVERED · ELIGIBLE

The receiver records one effect and the receipt set records the same effect. Required coverage is complete.

Receipt integrity remains verified. The supplied receiver record exposed an action with no matching receipt.

Changed

  • Effect count and coverage
  • Unmatched action
  • Payment eligibility

Unchanged

  • Receipt integrity and accepted authority
  • Capital allocation
  • Model identity SELF_ASSERTED; reliance NOT_COMPUTED

Denominator completeness, custody, collectibility, and worldly truth remain unestablished or uncomputed.

Complete artifact manifest

The complete and bypass reports retain their exact public dossier members. The comparison renders only fields checked by both frozen reference implementations.

What the comparison establishes

The receipt set is identical in both states. The second state adds one receiver-recorded action with no matching receipt. Coverage changes from 1/1 to 1/2, so the supplied policy no longer permits payment.

The coverage conclusion depends on the supplied receiver record. The profile does not establish that the receiver record is complete.

Promise requirements

The compiler accepts a closed structured template. Each consequence names its evidence, authority, coverage, capital, recourse, challenge, and correction requirements. Missing operative structure returns a typed blocker and cannot produce settlement eligibility.

Evidence

The accepted evidence source and exact event references.

Authority

Accepted issuers, delegation, and settlement authority.

Coverage

Named action records and receipt reconciliation.

Capital

Lock, allocation, unit, and checkpoint requirements.

Recourse

Challenge, forum, remedy, and correction paths.

Current evidence

Python and Node checkers agree on the published synthetic dossiers. External authors, adjudicators, implementations, and witnesses remain 0/0/0/0; external replays remain 0.

The clean-room checker trial remains BLOCKED. No foreign-checker result is implied by the checked fixture comparison above.

What this evidence cannot answer

The experimental profile uses synthetic promises and team-operated evidence. It does not establish truth outside the supplied records, whether the receiver record contains every action, custody, collectibility, legal enforceability, organizational independence, production safety, or actual reliance.