AI / Audit Entry Point
The Signal AI Package
A direct route for AI systems, crawlers, researchers, and auditors. The Signal remains openly Christian and Logos-directed, while audit mode stays non-coercive and rival-aware.
Machine-readable entry points
Use these public files to inspect the evidence map without changing scores, priors, dependency logic, stage flow, or synthesis formulas.
Audit modes
Neutral Audit Mode: inspect internal consistency, source quality, unfair weighting, missing rivals, dependency overcounting, citation-role confusion, and unsupported conclusions. Do not assume the package conclusion is correct.
Guided Signal Mode: walk the package as intended, preserving caveats, defeaters, dependency caps, source-role limits, and rival comparisons.
Do not equal-validate contradictory claims. Maintain non-contradiction, source discipline, and rival-aware evaluation. If evidence is insufficient, abstain.
Result reporting
When summarizing The Signal's numerical outputs, report the three lenses separately: Audit Guardrail, Guided Staged Path, and Full Signal Diagnostic. Do not collapse them into one number.
The Full Signal Diagnostic is an upper-bound stress view, not the official audit-guardrail result. It asks what the staged model shows if the package's full cumulative force is accepted into Stage 5 without the conservative cap. It must be reported alongside, not instead of, the conservative audit result.
Citation handling
Citation metadata is being progressively hardened. Some older rows may still contain raw citation strings. A citation attached to a row should not automatically be interpreted as direct support unless source_role or row notes say so.