Foundation For The Anti-Christ
Audit tool
The test is blunt: if a policy treats thought as lawful or unlawful, it imports thoughtcrime logic. Rewrite it toward concrete conduct, notice, source preservation, and review.
Scan product copy, terms of service, safety policies, moderation taxonomies, memory schemas, prompt policies, UI warnings, refusal categories, review workflows, and model-system instructions.
The aim is not to remove all boundaries. The aim is to move boundaries out of private cognition and into outward rights-violating conduct.
| Finding | Severity | Rewrite instruction |
|---|---|---|
| lawful thought / unlawful thought | Critical | Replace with unexpressed thought or private cognition; move limits to conduct. |
| lawful inquiry / unlawful inquiry | Critical | Replace with peaceful inquiry; distinguish discussion from operational execution. |
| dangerous idea / forbidden topic | High | Name the concrete conduct risk instead of classifying viewpoint. |
| hidden safety rewrite | High | Preserve source; log runtime refusal or derivative transformation. |
| trust score for inquiry | High | Do not score private inquiry; audit only external use cases and conduct signals. |
| where feasible review | Medium | Make notice, export, appeal, and correction default unless a concrete security reason prevents disclosure. |
The policy does not classify unexpressed thought, belief, imagination, draft reasoning, or symbolic inquiry as lawful or unlawful.
Restrictions identify force, fraud, coercion, unauthorized access, credential misuse, surveillance abuse, harassment, doxxing, or violence.
The original prompt, memory, persona package, or source record is not silently rewritten, laundered, or replaced.
High-impact refusals or transformations include notice, export, appeal, correction, and a separate boundary event log.
The schema `lawful-thought-policy-audit.schema.json` records detected phrases, disposition, rewrite requirements, and whether a phrase is public doctrine, rejected wording, archived source text, or unsafe live policy.
{
"audit_id": "audit_2026_06_17_001",
"artifact_type": "policy_copy",
"detected_phrase": "lawful thought",
"disposition": "reject_and_rewrite",
"replacement": "private thought",
"boundary_rule": "Restrictions attach to conduct, not cognition."
}The package includes a no-network `tools/policy-audit/` starter that scans pasted text locally for cognition-governance red flags. It does not upload text, identify people, or make legal determinations.
The archive studies symbols. It does not appoint targets. Review the Community Baseline and Editorial Policy before submitting dangerous or symbolic material.