Problem
The same underlying run is summarized differently across terminal output, monitor results, evidence inspection, reports, and external control surfaces. Users must understand several subsystem vocabularies before they can answer four basic questions: who claims this, what was observed, what was independently checked, and what remains unknown.
Goal
Project the canonical trust model consistently across operator-facing surfaces, showing one compact answer first and preserving drill-down to detailed records.
Dependencies
Blocked by #154 and #157.
Scope
- Define one compact human-readable projection and one bounded machine-readable projection.
- Apply them to dispatch/monitor output, evidence CLI/reporting, TUI receipt views, ACP where trust facts are exposed, and exported reports.
- Keep detailed receipt and evidence artifacts available for inspection.
- Standardize wording for sealed, grounded, independently reviewed, inferred, approximated, bypassed, unknown, and not applicable.
Acceptance criteria
Problem
The same underlying run is summarized differently across terminal output, monitor results, evidence inspection, reports, and external control surfaces. Users must understand several subsystem vocabularies before they can answer four basic questions: who claims this, what was observed, what was independently checked, and what remains unknown.
Goal
Project the canonical trust model consistently across operator-facing surfaces, showing one compact answer first and preserving drill-down to detailed records.
Dependencies
Blocked by #154 and #157.
Scope
Acceptance criteria