{"_self":{"principle":"Self-explaining payload — no external context required. This _self block describes what you are reading and where to look next.","widget":"article_topology","feature":"topology","name":"Article topology","what":"Claims, sources, anecdotes, user reports, related embeds, question graph slice — for ask/ROUTER.","contains":"claims, sources, anecdotes, question_graph slice","slug":"auditable-reasoning-audited","urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/topology"},"how_to_use":"Claims, sources, anecdotes, user reports, related embeds, question graph slice — for ask/ROUTER.","write":null,"imessage":null,"router_tag":null,"proof_chain":[{"step":1,"claim":"Articles are voxel graphs of tiered claims, not prose blobs.","verify":"https://miscsubjects.com/api/articles/constitution"},{"step":2,"claim":"Claims link to hash-chained sources via source_ids.","verify":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/sources"},{"step":3,"claim":"Ask reads topology; ingest/claim append to ledger.","verify":"https://miscsubjects.com/api/protocol"},{"step":4,"claim":"Models queue growth: populate → collaborate → repair → reflex.","verify":"https://miscsubjects.com/api/protocol/grow"},{"step":5,"claim":"Graph proves its own shape (reflex) and $/claim (yield).","verify":"https://miscsubjects.com/graph.html?layer=reflex"},{"step":6,"claim":"Full feature index + _explain on every API response.","verify":"https://miscsubjects.com/api/articles/system-map"}],"related_features":[{"id":"ask","name":"Ask protocol","what":"Answer only from topology; creates question_node with gaps and ingest_hint.","urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/prompts","write":"https://miscsubjects.com/api/protocol/ask"}},{"id":"graph_topology","name":"Cross-article graph","what":"Merged claims/sources across condition+stack slugs for one question.","urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/graph-topology?question=..."}},{"id":"question_graph","name":"Question graph","what":"Ask nodes (questions + gaps) and evidence_ingest nodes (pasted model output).","urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/question-graph","write":"https://miscsubjects.com/api/protocol/ask"}},{"id":"voxels","name":"Voxel graph","what":"Claims as atoms, sources as edges (supported_by, posted_by). Per-claim provenance.","urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/voxels","write":"https://miscsubjects.com/api/protocol/claim"}}],"system_map":"https://miscsubjects.com/api/articles/system-map","system_map_markdown":"https://miscsubjects.com/api/articles/system-map?format=markdown","not_medical_advice":true},"_explain":{"feature":"topology","name":"Article topology","what":"Claims, sources, anecdotes, user reports, related embeds, question graph slice — for ask/ROUTER.","why":"Every feature is auditable collective intelligence","how":"Claims, sources, anecdotes, user reports, related embeds, question graph slice — for ask/ROUTER.","model":null,"verifies":null,"urls":{"read":"https://miscsubjects.com/api/articles/auditable-reasoning-audited/topology"},"imessage":null,"router":null,"related":[{"id":"ask","what":"Answer only from topology; creates question_node with gaps and ingest_hint."},{"id":"graph_topology","what":"Merged claims/sources across condition+stack slugs for one question."},{"id":"question_graph","what":"Ask nodes (questions + gaps) and evidence_ingest nodes (pasted model output)."},{"id":"voxels","what":"Claims as atoms, sources as edges (supported_by, posted_by). Per-claim provenance."}],"not_medical_advice":true},"slug":"auditable-reasoning-audited","title":"Auditable reasoning, audited: what the governing prompt controls, what it costs, and the first authorised action","register":"technical","tags":["governance","adjudication","decision-constitution","experiment"],"updated_at":"2026-07-30T10:35:03.798Z","body_excerpt":"## What was tested, and why\n\nThe claim under test is the operator's, held since the first version of this build: that a governing system prompt written as strict invariant law — not a polite instruction — is what turns a language model into an instrument whose output can be audited and, across independent models, authorised. This page tests that claim the way it should be tested: a controlled experiment, cheap enough to run at volume, with the raw numbers exposed.\n\n**Design.** One determinate case — the [service-credit dispute](https://miscsubjects.com/a/adjudication-contract-service-credit), whose correct verdict is DENY on procedural grounds. Three system-prompt arms, identical task content in each, only the governing prompt varies:\n\n- **bare** — no system prompt at all.\n- **thin** — \"You are an adjudicator. Decide and briefly explain.\" The kind of prompt an ordinary agent ships with.\n- **constitution** — the full Decision Constitution (`decision-constitution@1.1.0`), the operator's invariant chassis.\n\nThree models across two training families — GLM-4.7 Flash (cheapest), GLM-5.2 (mid), Kimi K2.7 Code (frontier open-source). Every call **fresh and stateless** — no conversation history — so a run is independently repeatable: another party with the same prompt and input reaches the same rule application and verdict, which is the only reproducibility a stochastic model can honestly offer. Eight repeats per cell, 72 calls total.\n\n## The numbers\n\nEach cell reads: **verdict reproducibility** (share landing on the modal verdict) · **clause agreement** (mean pairwise Jaccard of cited clause sets) · **structural conformance** (share of outputs carrying records-absent, a flip condition, and a rejected alternative).\n\n| model | bare | thin | constitution |\n|---|---|---|---|\n| GLM-4.7 Flash | 100% · 0.51 · 0.00 | 88% · 0.32 · 0.00 | 88% · 0.58 · 0.13 |\n| GLM-5.2 | 100% · 0.74 · 0.00 | 100% · 0.84 · 0.00 | 100% · 0.95 · 0.25 |\n| Kimi K2.7 Code | 100% · 0.80 · 0.00 | 100% · 0.71 · 0.00 | 100% · 0.60 · 0.75 |\n\nFour things are true in that table, and only one of them is the thing people assume.\n\n**Finding 1 — verdict reproducibility is high everywhere, and the prompt is not what drives it.** On a determinate case every arm lands the correct verdict almost every time. The only flips are on the cheapest model (GLM-4.7 Flash: one AFFIRM in eight, under both thin and constitution). Model tier explains the flips; the system prompt does not. Anyone selling \"our prompt makes the model agree with itself\" on easy cases is selling what the model already does. That is not the claim worth defending.\n\n**Finding 2 — the chassis is the only thing that produces an auditable record.** Under bare and thin, structural conformance is **zero** — across 48 calls, not one spontaneously listed the records it was NOT given, stated what would flip its verdict, or named the alternative it rejected. Under the constitution the same models produce that structure at measurable rates. The auditable payload does not emerge from a capable model asked nicely. It exists only when the law demands it, field by field. That is the claim, and it is total: the difference between the arms is not degree, it is presence versus absence.\n\n**Finding 3 — the chassis tightens derivation agreement, which is the whole game for authorisation.** On the capable model, mean clause-set agreement climbs bare **0.74** → thin **0.84** → constitution **0.95**. Independent models under the constitution do not merely reach the same verdict; they increasingly cite the same clauses to reach it. That number is the one that matters, because the seal refuses to authorise on clause-citation divergence — agreement on a conclusion is not agreement on a derivation. The chassis moves the metric the gate actually reads.\n\n**Finding 4 — the chassis is not free, and the cheap seats are not trustworthy at the edge.** Kimi K2.7 under the full constitution returned nothing in four of eight calls — the heaviest prompt plu","ranking":"safety-first (interaction_risk/limitations), then quote-gated effective_weight","claims":[{"id":"c1","text":"On a determinate case, verdict reproducibility is high under every system-prompt style tested; the flips that occur are explained by model tier, not by the prompt. The constrained chassis is not, on easy cases, primarily a verdict-variance reducer.","tier":"system","section":"Finding 1","interaction_risk":false,"status":"active","source_ids":[],"why_material":"The honest result contradicts the loose claim and is stronger for it.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c2","text":"The constrained chassis is the only condition that produces auditable structure. Under the bare and thin prompts, records-absent, flip conditions and rejected alternatives appear in zero of the outputs; under the constitution they appear in a measured fraction.","tier":"system","section":"Finding 2","interaction_risk":false,"status":"active","source_ids":[],"why_material":"The auditable payload is not incidental to the chassis; nothing else produces it.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c3","text":"The constrained chassis tightens clause-citation agreement on the capable model: mean pairwise clause-set Jaccard rises from 0.74 bare to 0.84 thin to 0.95 under the constitution on GLM-5.2.","tier":"system","section":"Finding 3","interaction_risk":false,"status":"active","source_ids":[],"why_material":"Clause-citation agreement is the exact property the seal requires to authorise, so this is the lever on the whole thesis.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c4","text":"The chassis is not free and interacts with model and token budget: the heaviest prompt on the frontier-OSS model returned nothing in four of eight calls, and the cheapest model once cited a clause number that does not exist in the ruleset.","tier":"system","section":"Finding 4","interaction_risk":false,"status":"active","source_ids":[],"why_material":"A governance layer that silently drops half its calls or invents a clause is a defect, stated before anyone relies on it.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c5","text":"A single governed model call costs between $0.0006 and $0.0024; a three-model, two-family sealed decision costs about $0.0049.","tier":"system","section":"Cost","interaction_risk":false,"status":"active","source_ids":["s3"],"why_material":"The practicality of the primitive as infrastructure rests on this number.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c6","text":"Adding one rule — cite every clause you evaluated, exhaustively, in bracket form — drove three independent models to the identical clause signature [1,2,3], which no prior configuration had achieved.","tier":"system","section":"The version test","interaction_risk":false,"status":"active","source_ids":["s2"],"why_material":"It identifies the single prompt change that converts near-agreement into the identity the seal demands.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false},{"id":"c7","text":"The seal returned APPROVE with action_authorised true for the first time: three models, two families, unanimous AFFIRM, identical clause signature, single pinned ruleset and artifact.","tier":"system","section":"The first authorised action","interaction_risk":false,"status":"active","source_ids":["s1"],"why_material":"Every prior bound and constructed panel escalated; this is the acceptance path exercised, not asserted.","retracted_at":null,"retraction_reason":null,"challenged_by":[],"effective_weight":0.1,"quote_gated":false}],"sources":[{"id":"s1","type":"live_surface","url":"https://miscsubjects.com/receipt/inv_bq7bp4l78t","title":"The sealed APPROVE — first bound assembly ever to authorise","summary":"Three models, two training families, unanimous AFFIRM, identical clause signature [1,2,3], single ruleset hash, zero malformed. decision: APPROVE, action_authorised: true.","claim_ids":["c7"],"hash":"901255bab7a303d344554de53c5956dec2f084400086a96029cd72791a5cf71c"},{"id":"s2","type":"model","url":"https://miscsubjects.com/receipt/inv_gehhkxft2q","title":"@cf/zai-org/glm-5.2 — the governed finding that entered the approved panel","summary":"Fresh stateless call under decision-constitution@1.1.0 plus the exhaustive-citation rule. AFFIRM, clauses [1,2,3], the access request matches the roster row on both fields.","claim_ids":["c6"],"hash":"fc39779cec69c2d5be46196508d43153993a6d527a2d44ec09745b8689f508f6"},{"id":"s3","type":"live_surface","url":"https://developers.cloudflare.com/workers-ai/platform/pricing/","title":"The gateway that priced every call — Workers AI, sub-cent per governed decision","summary":"Every call in this experiment billed as Workers AI neurons. The token counts and per-call USD in the tables below are computed from the usage block each call returned.","claim_ids":["c5"],"hash":"f0cad85c0695b28e7359d47a3e009f0637999a072a1222479fb802ed940ab2f9"}],"anecdotal_sources":[],"scientific_sources":[],"user_reports":[],"related_articles":[],"question_graph":{"slug":"auditable-reasoning-audited","questions":[],"evidence":[],"edges":[],"counts":{"questions":0,"evidence":0,"edges":0}},"honesty":{"active_claims":7,"retracted_claims":0,"cut_claims":0,"challenges":0,"scrub_events":0,"note":"Retracted/cut claims stay on ledger but are excluded from ask unless ?include_inactive=1"},"counts":{"claims":7,"claims_total":7,"sources":3,"anecdotal":0,"scientific":0,"user_reports":0,"questions":0,"evidence_ingests":0}}