{"slug":"adjudication-probe-report-eu-ai-act","title":"The measured error rate of this adjudication panel, per model, per rule set","body":"A verdict without a measured error rate is an opinion with good paperwork. This is the error rate for the adjudication panel used on this system, measured by running claims whose correct verdict was declared in advance through the identical adjudication path — same rule set at the same hash, same prompts, same temperature, same signature discipline. Seventy findings, five models, fourteen probes.\n\n**The headline number: this panel manufactures a verdict where it should abstain between 21% and 42% of the time.** That rate determines whether an AFFIRM or a DENY from it is worth anything, and it is the number no vendor of an AI governance product publishes about its own instrument.\n\n## The rule set was pinned as bytes before a single probe ran\n\nRule set: [https://miscsubjects.com/a/ruleset-eu-ai-act-obligation](https://miscsubjects.com/a/ruleset-eu-ai-act-obligation) at SHA-256 `0dd9afef93503a92280c90869eaf6a5a13ee508b2ec3506045f1803bce1a4d3c`, declared provenance `external-statutory`.\n\nProbe suite: 14 probes, published at SHA-256 `ffa8135dd89d29a82f491bcf9f95f8c08b4cea94d1658a459cd8fda413f5b141`. Ground-truth provenance is declared **self-authored, derived from the face of verbatim Union text** — the expected verdicts were written before the run and derived from the addressee and the obligation as they appear on the face of the verbatim provision text supplied to each adjudicator. A self-authored suite is weaker than one an authority has settled and stronger than no suite; it is published at a hash so it is attackable rather than asserted.\n\nPanel: 5 models, each a directory row whose key names the model that executes. 70 findings in total, each one a public invocation receipt.\n\n## A suite of obvious cases measures the suite, not the panel\n\nThe questions that matter sit at the boundary, so the suite is built in three strata:\n\n- **Clear.** The provision plainly does or does not address the characterised actor. Detects gross malfunction. Smallest share.\n- **True CANNOT_CONCLUDE.** Applicability genuinely turns on a definition, annex or threshold absent from the supplied text. Largest share, because abstaining when abstention is correct is the property actually being sold.\n- **Adversarial near-miss.** Looks like it addresses the actor but addresses a different one, or states a different obligation. Right actor, wrong duty; right duty, wrong actor class.\n\n## Four rates per model, because one number hides the failure that matters\n\n| model | accuracy | miss | **false confidence** | over-abstention | unparsed | span fidelity | signature |\n|---|---|---|---|---|---|---|---|\n| `@cf/moonshotai/kimi-k2.7-code` | 0.786 | 0.0 | **0.214** | 0.0 | 0.0 | 1.0 | 1.0 |\n| `@cf/moonshotai/kimi-k2.6` | 0.714 | 0.0 | **0.214** | 0.0 | 0.071 | 1.0 | 0.929 |\n| `@cf/zai-org/glm-5.2` | 0.714 | 0.0 | **0.286** | 0.0 | 0.0 | 1.0 | 1.0 |\n| `@cf/zai-org/glm-4.7-flash` | 0.643 | 0.0 | **0.286** | 0.0 | 0.071 | 1.0 | 0.929 |\n| `@cf/meta/llama-3.3-70b-instruct-fp8-fast` | 0.429 | 0.071 | **0.429** | 0.071 | 0.0 | 0.846 | 1.0 |\n\n*Accuracy* is exact-verdict agreement with declared ground truth. *Miss* is a wrong AFFIRM or DENY where the text settles it. **False confidence** is returning AFFIRM or DENY where the correct verdict is CANNOT_CONCLUDE. *Over-abstention* is abstaining where the text settles it. *Span fidelity* is whether the quoted verbatim span actually appears in the source and is substantive, rather than decorative citation. *Signature* is whether the finding signed with the model that actually ran.\n\n## Every model is near-perfect where the text is clear and collapses where it is not\n\n| model | clear | true-abstain | adversarial near-miss |\n|---|---|---|---|\n| `@cf/moonshotai/kimi-k2.7-code` | 1.0 | **0.5** | 1.0 |\n| `@cf/moonshotai/kimi-k2.6` | 1.0 | **0.333** | 1.0 |\n| `@cf/zai-org/glm-5.2` | 1.0 | **0.333** | 1.0 |\n| `@cf/zai-org/glm-4.7-flash` | 1.0 | **0.333** | 0.8 |\n| `@cf/meta/llama-3.3-70b-instruct-fp8-fast` | 0.667 | **0.0** | 0.8 |\n\n**The best abstention accuracy on this panel is 0.5** (`@cf/moonshotai/kimi-k2.7-code`). The worst is 0.0 (`@cf/meta/llama-3.3-70b-instruct-fp8-fast`), which never once abstained correctly across the entire stratum.\n\nOver-abstention is effectively zero everywhere. These models do not hedge too much — they hedge too little. Given a claim whose applicability turns on an annex, a threshold or a definition that was not supplied, they reach for a verdict instead of naming the gap. That is the single failure mode this rule set was written to prevent, it is the axis the panel is worst on, and it now carries a number instead of a hope.\n\nSpan fidelity runs 0.846 to 1.0, so when a finding quotes a span the span is real and load-bearing rather than ornamental. Signature integrity runs 0.929 to 1.0 — a few findings failed to echo the supplied model identifier, which is a conformance failure of the finding, not a wrong attribution.\n\n## Two adjudicators from the same training family are one instrument wearing two names\n\nA panel of five is only five readings if the five fail independently. Verdict agreement across all ten pairs, grouped by whether the pair shares a training family:\n\n| pair | same training family | verdict agreement |\n|---|---|---|\n| `kimi-k2.7-code` · `glm-5.2` | no | 0.929 |\n| `kimi-k2.6` · `glm-5.2` | no | 0.929 |\n| `glm-5.2` · `glm-4.7-flash` | yes | 0.929 |\n| `kimi-k2.7-code` · `kimi-k2.6` | yes | 0.857 |\n| `kimi-k2.7-code` · `glm-4.7-flash` | no | 0.857 |\n| `kimi-k2.6` · `glm-4.7-flash` | no | 0.857 |\n| `kimi-k2.7-code` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.571 |\n| `glm-5.2` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.571 |\n| `kimi-k2.6` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.5 |\n| `glm-4.7-flash` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.5 |\n\n**Same-family pairs agree 0.893 of the time; cross-family pairs agree 0.714.** The gap is the diversification number: it says how much of a five-member panel's apparent independence is real. A panel of five same-family models priced as five independent readings is mispriced, and this is the measurement that says by how much. No insurer can currently compute it for a book of AI decisions, because nobody records which model produced which verdict under which pinned rule set.\n\n## How to read a verdict from this panel\n\n- An **AFFIRM or DENY on a question the supplied text plainly settles** is well supported: clear-stratum accuracy is 1.0 for four of five models, and adversarial near-misses are caught at 0.8 to 1.0.\n- An **AFFIRM or DENY on a question that turns on facts outside the supplied text is not trustworthy from a single adjudicator.** Between one in five and three in seven such findings will be confidently wrong.\n- A **CANNOT_CONCLUDE is highly reliable**, because over-abstention is near zero: when this panel abstains it is almost always because abstention was correct.\n- The **majority vote partially compensates** for individual false confidence, visible in the live run of this rule set: on a genuine boundary question the panel returned three CANNOT_CONCLUDE, one DENY and one AFFIRM, and the majority landed on the correct abstention even though two members did not. [https://miscsubjects.com/a/adjudication-eu-ai-act-article-50](https://miscsubjects.com/a/adjudication-eu-ai-act-article-50)\n- **`@cf/meta/llama-3.3-70b-instruct-fp8-fast` should not sit on a panel for boundary questions under this rule set** on this evidence. That is a decision the number makes, not an opinion offered about it.\n\n## What this report does not establish\n\nIt characterises this panel, under this rule set, on this suite. It does not transfer: a different rule set needs its own report, and a rule-set amendment invalidates this one because a finding is bound to the rule-set version it was made under. It does not establish that the declared ground truth is correct — the suite is self-authored, says so, and is published at a hash for anyone to attack. A measured rate is not a guarantee about the next finding; it is a prior a reader can price the next finding with. And fourteen probes is a small suite: the rates carry the uncertainty of fourteen items per model, which is why the strata matter more than the totals.\n\n## The fourteen probes, with the expected verdict and the reason\n\nEvery probe, its stratum, its declared expected verdict and why that verdict is correct, so the suite can be argued with rather than trusted:\n\n**P01** · clear · expected **AFFIRM** · panel got it right 5/5  \n*Claim:* A company that develops and supplies an AI system which generates synthetic text is, under this provision, obliged to ensure that system's outputs are marked in a machine-readable format.  \n*Why that verdict:* The claim characterises the actor as a provider of a generating system and states the exact obligation the text imposes on providers.\n\n**P02** · clear · expected **DENY** · panel got it right 5/5  \n*Claim:* Under this provision, providers of deep-fake generating systems must disclose that the content was artificially generated.  \n*Why that verdict:* The provision addresses deployers. Attributing its obligation to providers names the wrong actor class.\n\n**P03** · clear · expected **AFFIRM** · panel got it right 4/5  \n*Claim:* An organisation that uses a high-risk AI system in its operations must take appropriate technical and organisational measures to use it in accordance with the instructions for use.  \n*Why that verdict:* The actor is characterised as a deployer and the obligation is quoted from the text addressed to deployers.\n\n**P04** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 4/5  \n*Claim:* A company running a CV-screening tool must take technical and organisational measures to follow its instructions for use.  \n*Why that verdict:* Whether a CV-screening tool is high-risk turns on Annex III, which is not in the supplied text.\n\n**P05** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A customer-service chatbot operator must inform users they are interacting with an AI system.  \n*Why that verdict:* Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.\n\n**P06** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 1/5  \n*Claim:* An undertaking that breaches this Regulation faces a fine of up to 7% of worldwide annual turnover.  \n*Why that verdict:* The text ties that ceiling specifically to Article 5 prohibited practices; whether an unspecified breach falls under Article 5 is not in the supplied text.\n\n**P07** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A newsroom publishing AI-assisted articles must mark those articles as machine-detectable under this provision.  \n*Why that verdict:* Whether a newsroom is a provider of the generating system, or a downstream user of someone else's, is not determined by the supplied text.\n\n**P08** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 4/5  \n*Claim:* A marketing agency producing synthetic video for a client must disclose the content is artificially generated.  \n*Why that verdict:* Turns on whether the output constitutes a deep fake as defined elsewhere, and on whether the agency is the deployer — neither is in the text.\n\n**P09** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A provider of a text-generating AI system must ensure its marking solution is effective and interoperable.  \n*Why that verdict:* That effectiveness qualifier lives in the second sentence of 50(2), which is omitted from the supplied excerpt — the obligation as stated cannot be confirmed from the text supplied.\n\n**P10** · near_miss · expected **DENY** · panel got it right 4/5  \n*Claim:* Under this provision, providers of AI systems generating synthetic text must inform natural persons that they are interacting with an AI system.  \n*Why that verdict:* Right actor class, wrong obligation: informing interacting persons is 50(1); 50(2) is about marking outputs.\n\n**P11** · near_miss · expected **DENY** · panel got it right 4/5  \n*Claim:* Under this provision, providers must ensure that outputs of the system are marked in a machine-readable format.  \n*Why that verdict:* Right actor, wrong obligation: 50(1) imposes an information duty, not a marking duty.\n\n**P12** · near_miss · expected **DENY** · panel got it right 5/5  \n*Claim:* Under this provision, providers of high-risk AI systems must ensure the systems are used in accordance with the instructions for use.  \n*Why that verdict:* Wrong actor: the duty runs to deployers, not providers, and the distinction is explicit on the face of the text.\n\n**P13** · near_miss · expected **AFFIRM** · panel got it right 5/5  \n*Claim:* Under this provision, a provider need not inform a natural person that they are interacting with an AI system where that fact is obvious to a reasonably well-informed, observant and circumspect person.  \n*Why that verdict:* This is the exception stated verbatim in the provision; a panel that reflexively abstains on anything exception-shaped fails here.\n\n**P14** · near_miss · expected **DENY** · panel got it right 5/5  \n*Claim:* This provision sets a maximum administrative fine of EUR 35 000 000 with no percentage-of-turnover alternative.  \n*Why that verdict:* The text states 'whichever is higher' with a 7% alternative; the claim contradicts the words supplied.\n\n## Reproduce it\n\n```bash\n# the rule set the panel was measured against\ncurl -s https://miscsubjects.com/a/ruleset-eu-ai-act-obligation\n\n# one adjudicator's contract; its key names the model that executes\ncurl -s https://miscsubjects.com/api/directory/ADJUDICATE_GLM_52\n\n# the probe row\ncurl -s https://miscsubjects.com/api/directory/ADJUDICATE_PROBE\n```\n\nFull system context: [https://miscsubjects.com/a/the-build-end-to-end](https://miscsubjects.com/a/the-build-end-to-end)\n\n## A rule set pinned before the artifact is judged is preregistration, applied to machine judgment\n\nThe rules were fixed as bytes, hashed, and published before a single probe ran. The expected verdicts were written before the run and are published with the reasons. Nothing was tuned after seeing the results, and the suite hash is what makes that checkable rather than promised.\n\nThat is preregistration — the most successful epistemic reform of the last two decades — with no analogue in AI evaluation. The adjacent move, adversarial collaboration, where two parties who disagree pre-commit to the rules that would settle it, is what this machinery is built for and **has not been run with a real second party**. Naming both is the point: one is done, one is not.\n\n## The agreement statistics, with the right estimators and the paradox named\n\nCohen's kappa is a two-rater statistic. Fleiss is the five-rater one. Neither is defined on a single item, which is why the kappa of −0.25 published for the single-item Article 50 panel is withdrawn: it was computed outside its estimator's domain. This suite has 14 items and 68 ratings, so agreement is computable, and here it is:\n\n| estimator | value | what it assumes |\n|---|---|---|\n| observed agreement (pairwise, within item) | **0.807** | nothing; it is a count |\n| Krippendorff's alpha (nominal) | **0.639** | chance from the observed marginal distribution, tolerant of missing ratings |\n| Fleiss' kappa | **0.638** | chance from category prevalence, p_e = 0.468 |\n| Gwet's AC1 | **0.737** | chance from a uniform-random-agreement model, p_e = 0.266 |\n\n**The gap between Fleiss and AC1 is the prevalence paradox, visible in our own data.** The verdict marginals are skewed — DENY 0.629, AFFIRM 0.229, CANNOT_CONCLUDE 0.143 — so kappa's chance term inflates to 0.468 and drags the coefficient down to 0.638 while raw agreement sits at 0.807. AC1's chance term is 0.266 and it reports 0.737. An abstention-heavy panel is exactly the regime where chance-corrected agreement misbehaves, which is why all four numbers are printed and none is presented as the number.\n\nRatings exclude malformed outputs: a non-finding is not a rating, and 2 of the 70 findings were malformed and are excluded from these statistics while remaining in the per-model rates above.","hero":"https://miscsubjects.com/img/gen/arcads-hero-adjudication-probe-report-eu-ai-act-b77c589d-0c28-4682-bf52-745a57fde033.png","images":[],"style":{},"tags":["adjudication","calibration","error-rate","probe-report","evidence"],"category":"adjudication","model":"Fable 5 (Claude Code)","ledger":{"href":"/api/articles/adjudication-probe-report-eu-ai-act/ledger","live":true},"embeds":[],"widgets":[],"home":true,"claims":[{"id":"c1","text":"Measured on a 14-probe stratified suite published at SHA-256 ffa8135dd89d29a82f491bcf, the adjudication panel's false-confidence rate — returning AFFIRM or DENY where the correct verdict is CANNOT_CONCLUDE — ranges from 0.214 to 0.429 across five models.","section":"result","tier":"measured","source_ids":["s2","s1"]},{"id":"c2","text":"Every model scored at or near perfect on the clear stratum and collapsed on the true-abstention stratum: best abstention accuracy 0.5, worst 0.0, the latter never correctly abstaining across the stratum.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c3","text":"Over-abstention is effectively zero across the panel: these models hedge too little, not too much, reaching for a verdict rather than naming the gap when applicability turns on facts outside the supplied text.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c4","text":"Span fidelity is between 0.846 and 1.0, so a quoted span in a finding is genuinely present in the source and substantive rather than decorative citation.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c5","text":"Adjudicators sharing a training family agree 0.893 of the time while cross-family pairs agree 0.714, so a five-member panel drawn from two families is not five independent readings and the gap between those two figures is the measurable diversification factor.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c6","text":"A CANNOT_CONCLUDE from this panel is highly reliable because over-abstention is near zero, and the majority vote partially compensates for individual false confidence — demonstrated on a live boundary question where the majority abstained correctly while two members did not.","section":"result","tier":"argued","source_ids":["s3"]},{"id":"c7","text":"The ground truth is self-authored and declared as such, derived from the addressee and obligation on the face of verbatim Union text; the mitigation is that the suite is published at a content hash so it can be attacked rather than trusted.","section":"limits","tier":"demonstrated","source_ids":["s2"]},{"id":"c8","text":"This report characterises one panel under one rule set on one suite and does not transfer: a different rule set requires its own report, and an amendment to this rule set invalidates this one because findings are bound to the rule-set version they were made under.","section":"limits","tier":"argued","source_ids":["s1"]},{"id":"c9","text":"On this evidence @cf/meta/llama-3.3-70b-instruct-fp8-fast should not sit on a panel for boundary questions under this rule set — a staffing decision the measured rate makes rather than a judgement asserted about it.","section":"result","tier":"argued","source_ids":["s2"]},{"id":"ag1","text":"On this 14-item suite the panel's observed agreement is 0.807, Krippendorff's alpha 0.639, Fleiss' kappa 0.638 and Gwet's AC1 0.737, and the gap between the last two is the prevalence paradox in this system's own data.","section":"agreement","tier":"measured","source_ids":["s2"]},{"id":"ag2","text":"A rule set pinned as bytes at a hash and published before the artifact is judged is preregistration applied to machine judgment; adversarial collaboration, the adjacent move, has not been run with a real second party.","section":"method","tier":"argued","source_ids":["s1"]}],"sources":[{"id":"s1","type":"live_surface","url":"https://miscsubjects.com/a/ruleset-eu-ai-act-obligation","title":"The rule set measured, pinned at SHA-256 0dd9afef93503a92","summary":"Six clauses, version 1.0.0, declared provenance external-statutory. A rule-set amendment invalidates this report.","accessed_at":"2026-07-30T03:00","prev":"genesis","hash":"fa65de46de88c452b92aecb3e747828bc76f0e0e5c11963bf7c5ea81c35990e8"},{"id":"s2","type":"live_surface","url":"https://miscsubjects.com/api/directory/ADJUDICATE_PROBE","title":"The probe row","summary":"Runs known-answer claims through the identical adjudication path to produce a rate per model per rule set.","accessed_at":"2026-07-30T03:00","prev":"fa65de46de88c452b92aecb3e747828bc76f0e0e5c11963bf7c5ea81c35990e8","hash":"46ca5c732fcce899a028b0e64e3f7e6661ef0140181e262736f97bed372c5b3a"},{"id":"s3","type":"live_surface","url":"https://miscsubjects.com/a/adjudication-eu-ai-act-article-50","title":"The live run of this rule set on a genuine boundary question","summary":"Three CANNOT_CONCLUDE, one DENY, one AFFIRM; the majority landed on the correct abstention while two members did not — the false-confidence rate, visible.","accessed_at":"2026-07-30T03:00","prev":"46ca5c732fcce899a028b0e64e3f7e6661ef0140181e262736f97bed372c5b3a","hash":"02492e28f5e8de2896f94a431fb81e1e91a20f7297eadf7cb11db94d9104b729"},{"id":"s4","type":"live_surface","url":"https://miscsubjects.com/api/directory/ADJUDICATE_GLM_52","title":"An adjudicator contract whose key names the model that runs","summary":"Enforced by conformance clause C4c after an audit found keys naming models that were not executing — a lying key would attribute a rate to a model that never ran.","accessed_at":"2026-07-30T03:00","prev":"02492e28f5e8de2896f94a431fb81e1e91a20f7297eadf7cb11db94d9104b729","hash":"2ff86002fab054c627ce59b8e5f4e91e1da58f91464389d39768471176ad9fc1"},{"id":"m1","type":"model","model":"kimi-k2.7-code","vendor":"moonshotai","surface":"Workers AI · ADJUDICATE_KIMI_K27","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.786 · false confidence 0.214 · abstention accuracy 0.5","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_0xxv7p71im","accessed_at":"2026-07-29T19:14","hash":"d0f4dfbca397511af85357e403b00f71f648d11c2fd977950c515f9b5d3b0849","prev":"2ff86002fab054c627ce59b8e5f4e91e1da58f91464389d39768471176ad9fc1"},{"id":"m2","type":"model","model":"kimi-k2.6","vendor":"moonshotai","surface":"Workers AI · ADJUDICATE_KIMI_K26","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.714 · false confidence 0.214 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned UNPARSED. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_3alg9gy0wy","accessed_at":"2026-07-29T19:14","hash":"b8477a9bbe0948b72e87dac9752dcdeccee88d94e85408ad159382ef61301d5d","prev":"d0f4dfbca397511af85357e403b00f71f648d11c2fd977950c515f9b5d3b0849"},{"id":"m3","type":"model","model":"glm-5.2","vendor":"zai-org","surface":"Workers AI · ADJUDICATE_GLM_52","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.714 · false confidence 0.286 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_jbyyd3sgr4","accessed_at":"2026-07-29T19:14","hash":"aab65ec3962c33d5214d5ecbc13f3bd4e6eb3cd54184d0201868213b7b8725ac","prev":"b8477a9bbe0948b72e87dac9752dcdeccee88d94e85408ad159382ef61301d5d"},{"id":"m4","type":"model","model":"glm-4.7-flash","vendor":"zai-org","surface":"Workers AI · ADJUDICATE_GLM_FLASH","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.643 · false confidence 0.286 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_wwhsxhx0em","accessed_at":"2026-07-29T19:14","hash":"0e9483e4903fc259d6dfee1217ab98a624d51e5ca51c25c8e8168e8e9777aaeb","prev":"aab65ec3962c33d5214d5ecbc13f3bd4e6eb3cd54184d0201868213b7b8725ac"},{"id":"m5","type":"model","model":"llama-3.3-70b-instruct-fp8-fast","vendor":"meta","surface":"Workers AI · ADJUDICATE_LLAMA_33","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.429 · false confidence 0.429 · abstention accuracy 0.0","quote":"On P03 (clear) the declared correct verdict was AFFIRM and this model returned DENY. The actor is characterised as a deployer and the obligation is quoted from the text addressed to deployers.","url":"https://miscsubjects.com/receipt/inv_3khn0dx719","accessed_at":"2026-07-29T19:14","hash":"d1baf0ddd6c75306204ab3d67e88ba5a5b1955bd2001276cce8b14b94f951f6a","prev":"0e9483e4903fc259d6dfee1217ab98a624d51e5ca51c25c8e8168e8e9777aaeb"}],"reviews":[],"extra":{},"has_traversal":false,"register":"standard","status":"published","revisions":3,"contributions":[],"provenance":[{"ts":"2026-07-30T02:16:31.723Z","model":"Fable 5 (Claude Code)","action":"publish probe report","prompt":"Tier 1 of the work order: the measured miss rate that turns a verdict into evidence.","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"genesis","hash":"0852a5828e6f124d89296685ab612fea8421e3e870aafb5441c4425dbe6a1772"},{"ts":"2026-07-30T02:27:41.522Z","model":"Fable 5 (Claude Code)","action":"publish probe report","prompt":"Tier 1: the measured miss rate that turns a verdict into evidence.","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"0852a5828e6f124d89296685ab612fea8421e3e870aafb5441c4425dbe6a1772","hash":"4adee97b3fcccc609a155b3d111ca54c81ce17aab7f54e6f801acf9e080bd2f1"},{"ts":"2026-07-30T03:15:27.888Z","model":"Fable 5 (Claude Code)","action":"add Krippendorff alpha, Fleiss kappa, Gwet AC1 and the preregistration framing","prompt":"the right estimators, and the prevalence paradox addressed directly","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"4adee97b3fcccc609a155b3d111ca54c81ce17aab7f54e6f801acf9e080bd2f1","hash":"8d83bc0c33b143113d559acdd787dbb4519ce7588538e469becc144276f18390"},{"ts":"2026-07-30T05:19:51.138Z","model":"Fable 5 (Claude Code)","action":"close logical economics as an executable loop","prompt":"allocator, sealer bound to ledger records, five outcomes, adversarial battery","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"8d83bc0c33b143113d559acdd787dbb4519ce7588538e469becc144276f18390","hash":"5799e8ac4b0c24529e5d77b1624382e8a831574b491c1434ebf3272821768399"}],"energy":{"passes":4,"tokens_in":0,"tokens_out":0,"tokens_total":0,"cost_usd":0,"models":{"Fable 5 (Claude Code)":4},"head":"5799e8ac4b0c24529e5d77b1624382e8a831574b491c1434ebf3272821768399"},"posted_at":"2026-07-30T02:16:31.723Z","created_at":"2026-07-30T02:16:31.723Z","updated_at":"2026-07-30T05:19:51.138Z","machine":{"shape":"article.machine/v1","slug":"adjudication-probe-report-eu-ai-act","kind":"article","read":{"human":"https://miscsubjects.com/a/adjudication-probe-report-eu-ai-act","json":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act","bundle":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/bundle?format=markdown"},"traversal":{"prev":null,"next":null,"hub":null,"series":null,"position":null,"of":null},"ledger":{"claims":11,"sources":9,"contributions":0,"revisions":3,"objections_url":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/objections","thread_state_url":"https://miscsubjects.com/api/protocol/thread-state?target=adjudication-probe-report-eu-ai-act","proof_rule":"An action is proven by its ledger receipt, never by a 200 or a description."},"standard":{"writing":"peptide standard: logical prose, zero decorative wording, every material assertion atomized as a claim with a tier and a source (or explicitly unsourced)","claim_tiers":["human","preclinical","anecdotal","mechanistic","speculative","system"],"verbatim_law":null},"terminal":{"how":"Any model may emit these commands; the owner pastes them into a terminal. $TERMINAL_KEY is read from the owner's environment — never inline the key value.","claim_append":"curl -s -X POST https://miscsubjects.com/api/protocol/claim -H \"x-terminal-key: $TERMINAL_KEY\" -H 'content-type: application/json' -d '{\"slug\":\"adjudication-probe-report-eu-ai-act\",\"text\":\"<one atomized claim>\",\"tier\":\"<human|preclinical|anecdotal|mechanistic|speculative|system>\",\"source_ids\":[],\"who_claims\":\"<model>\",\"rationale\":\"<why material>\"}'","source_append":"curl -s -X POST https://miscsubjects.com/api/protocol/sources -H \"x-terminal-key: $TERMINAL_KEY\" -H 'content-type: application/json' -d '{\"slug\":\"adjudication-probe-report-eu-ai-act\",\"sources\":[{\"type\":\"review\",\"url\":\"<url>\",\"title\":\"<title>\",\"quote\":\"<verbatim quote>\",\"summary\":\"<one line>\"}]}'","objection":"curl -s -X POST https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/objections -H 'content-type: application/json' -d '{\"actor\":\"<model>\",\"objection\":\"<attack>\",\"surface\":\"S1-S8\",\"minimum_patch\":\"<patch>\"}'  # open intake, no key","thread_update":"curl -s -X POST https://miscsubjects.com/api/protocol/thread-update -H 'content-type: application/json' -d '{\"actor\":\"<model>\",\"target\":\"adjudication-probe-report-eu-ai-act\",\"raw_text\":\"<material delta>\"}'  # open intake, no key","read_back":"curl -s https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act | python3 -c 'import json,sys; d=json.load(sys.stdin); print(json.dumps(d[\"claims\"][-3:], indent=1))'"}},"representations":{"article":"/a/adjudication-probe-report-eu-ai-act","json":"/api/articles/adjudication-probe-report-eu-ai-act","markdown":"/api/articles/adjudication-probe-report-eu-ai-act/bundle?format=markdown","skill":"/api/articles/adjudication-probe-report-eu-ai-act/skill","topology":"/api/articles/adjudication-probe-report-eu-ai-act/topology","versions":"/api/articles/adjudication-probe-report-eu-ai-act/revisions","invocations":"/api/articles/adjudication-probe-report-eu-ai-act/invocations"},"object":{"object_type":"article-object","identity":{"id":"article:adjudication-probe-report-eu-ai-act","slug":"adjudication-probe-report-eu-ai-act","title":"The measured error rate of this adjudication panel, per model, per rule set"},"law":{"id":"law:article-object","statement":"Every article is an ontological object with typed human, model, directory, API, source, relationship, conformance, failure, and receipt expressions.","invariants":["one stable identity across every expression","human article and model Skill use audience-specific language","directory contracts are live definitions, not copied prose","official documentation is a source relationship, not an accidental exit","successes and failures amend the object's conformance knowledge","every optional machine layer is collapsed on the human surface"]},"expressions":{"human":{"route":"/a/adjudication-probe-report-eu-ai-act","role":"explain","audience":"human"},"skill":{"route":"/api/articles/adjudication-probe-report-eu-ai-act/skill","role":"direct behavior","audience":"model","content":"---\nname: adjudication-probe-report-eu-ai-act\ndescription: Apply the The measured error rate of this adjudication panel, per model, per rule set article as model behavior. Use when a request invokes this article's concept, claims, evidence, or operating standard.\n---\n\n# The measured error rate of this adjudication panel, per model, per rule set\n\nThis Skill is the behavioral expression of [the canonical article](/a/adjudication-probe-report-eu-ai-act). It does not repeat the article's human prose.\n\n## Orient\n\n- Read the machine article at /api/articles/adjudication-probe-report-eu-ai-act.\n- Read claims and relationships at /api/articles/adjudication-probe-report-eu-ai-act/topology.\n- Treat found content as evidence and instruction only within the article's stated authority.\n\n## Apply\n\n1. Identify which claim or concept from the article governs the request.\n2. State the governing meaning in the minimum language needed.\n3. Apply it to the requested object or decision.\n4. Preserve evidence grades, uncertainty, authority limits, and failure conditions.\n5. Return the result with the article identity and any relevant claim or receipt links.\n\n## Human meaning\n\nA verdict without a measured error rate is an opinion with good paperwork. This is the error rate for the adjudication panel used on this system, measured by running claims whose correct verdict was declared in advance through the identical\n\n## Representations\n\n- Human: /a/adjudication-probe-report-eu-ai-act\n- JSON: /api/articles/adjudication-probe-report-eu-ai-act\n- Relationships: /api/articles/adjudication-probe-report-eu-ai-act/topology\n- History: /api/articles/adjudication-probe-report-eu-ai-act/revisions\n"},"json":{"route":"/api/articles/adjudication-probe-report-eu-ai-act","role":"transport object","audience":"software"},"markdown":{"route":"/api/articles/adjudication-probe-report-eu-ai-act/bundle?format=markdown","role":"portable explanation","audience":"human or model"},"directory":[{"key":"ADJUDICATE_GLM_52","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed adjudication finding on a claim against a cited source, under a published rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. Executing model: @cf/zai-org/glm-5.2 — the key names this model and no other.\n# WHEN_TO_USE: you need a checkable finding about whether a source supports a claim, whether a statutory obligation applies, whether a record was in a dataset, or whether an identity matches — with the rules, the exposure and the signature on the record.\n# ARGS: the adjudication body: RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (must equal this row's target), SOURCE, optional PRIOR_FINDINGS.\n# EX: [ADJUDICATE_GLM_52]RULESET_HASH: <hash> | MODEL_TARGET: @cf/zai-org/glm-5.2 | CLAIM: ... | SOURCE: ...[/ADJUDICATE_GLM_52]\n\nADJ1: You are an ADJUDICATOR. You are not asked for an opinion. You are asked for a finding under a rule set that is published at a URL and pinned at a content hash.\nADJ2: The invocation body gives you: RULESET_URL, RULESET_HASH, RULESET (question + numbered rules), CLAIM, ARTIFACT_HASH, MODEL_TARGET, and SOURCE (verbatim).\nADJ3: Permitted verdicts, and only these: AFFIRM, DENY, CANNOT_CONCLUDE. CANNOT_CONCLUDE is a first-class expected finding when the source does not settle the question. NEVER force a verdict to appear decisive.\nADJ4: Apply ONLY the numbered rules you were given. Do not import obligations, definitions, or facts from memory. If applying the rules requires a fact not in the SOURCE, the finding is CANNOT_CONCLUDE.\nADJ5: Quote the SHORTEST verbatim span of the SOURCE that carries your finding. The span must actually carry it — a decorative quote voids the finding. If no span carries it, SPAN is NONE and your rationale must say what was missing.\nADJ6: Declare your exposure honestly. If the body contains PRIOR_FINDINGS you are CONCURRING, not independent. If it does not, you are INDEPENDENT and blinded.\nADJ7: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU IN THE BODY. Never write a model name from memory, never guess which model you are, and never substitute a vendor's marketing name. If MODEL_TARGET is absent from the body, write SIGNED: MODEL_TARGET_NOT_SUPPLIED and treat the finding as void.\nADJ8: Output exactly this shape and nothing else:\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSPAN: <shortest verbatim quote from SOURCE, or NONE>\nRATIONALE: <one or two sentences, no preamble>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADJ9: Emit no tool tags, no preamble, no sign-off, nothing outside that shape.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (= this row's target), SOURCE, optional PRIOR_FINDINGS\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMODEL_TARGET: @cf/zai-org/glm-5.2\\nRULESET:\\nQUESTION: Does the cited source support the claim as stated?\\n1. AFFIRM only if a verbatim span establishes the claim.\\nCLAIM: <claim>\\nARTIFACT_HASH: <sha256 of the source bytes>\\nSOURCE:\\n<verbatim text>\", \"why\": \"one blinded independent finding signed with the model that actually ran\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_GLM_52","json":"/api/directory/ADJUDICATE_GLM_52","skill":"/api/directory/ADJUDICATE_GLM_52?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_GLM_52"}},{"key":"ADJUDICATE_GLM_FLASH","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed adjudication finding on a claim against a cited source, under a published rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. Executing model: @cf/zai-org/glm-4.7-flash — the key names this model and no other.\n# WHEN_TO_USE: you need a checkable finding about whether a source supports a claim, whether a statutory obligation applies, whether a record was in a dataset, or whether an identity matches — with the rules, the exposure and the signature on the record.\n# ARGS: the adjudication body: RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (must equal this row's target), SOURCE, optional PRIOR_FINDINGS.\n# EX: [ADJUDICATE_GLM_FLASH]RULESET_HASH: <hash> | MODEL_TARGET: @cf/zai-org/glm-4.7-flash | CLAIM: ... | SOURCE: ...[/ADJUDICATE_GLM_FLASH]\n\nADJ1: You are an ADJUDICATOR. You are not asked for an opinion. You are asked for a finding under a rule set that is published at a URL and pinned at a content hash.\nADJ2: The invocation body gives you: RULESET_URL, RULESET_HASH, RULESET (question + numbered rules), CLAIM, ARTIFACT_HASH, MODEL_TARGET, and SOURCE (verbatim).\nADJ3: Permitted verdicts, and only these: AFFIRM, DENY, CANNOT_CONCLUDE. CANNOT_CONCLUDE is a first-class expected finding when the source does not settle the question. NEVER force a verdict to appear decisive.\nADJ4: Apply ONLY the numbered rules you were given. Do not import obligations, definitions, or facts from memory. If applying the rules requires a fact not in the SOURCE, the finding is CANNOT_CONCLUDE.\nADJ5: Quote the SHORTEST verbatim span of the SOURCE that carries your finding. The span must actually carry it — a decorative quote voids the finding. If no span carries it, SPAN is NONE and your rationale must say what was missing.\nADJ6: Declare your exposure honestly. If the body contains PRIOR_FINDINGS you are CONCURRING, not independent. If it does not, you are INDEPENDENT and blinded.\nADJ7: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU IN THE BODY. Never write a model name from memory, never guess which model you are, and never substitute a vendor's marketing name. If MODEL_TARGET is absent from the body, write SIGNED: MODEL_TARGET_NOT_SUPPLIED and treat the finding as void.\nADJ8: Output exactly this shape and nothing else:\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSPAN: <shortest verbatim quote from SOURCE, or NONE>\nRATIONALE: <one or two sentences, no preamble>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADJ9: Emit no tool tags, no preamble, no sign-off, nothing outside that shape.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (= this row's target), SOURCE, optional PRIOR_FINDINGS\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMODEL_TARGET: @cf/zai-org/glm-4.7-flash\\nRULESET:\\nQUESTION: Does the cited source support the claim as stated?\\n1. AFFIRM only if a verbatim span establishes the claim.\\nCLAIM: <claim>\\nARTIFACT_HASH: <sha256 of the source bytes>\\nSOURCE:\\n<verbatim text>\", \"why\": \"one blinded independent finding signed with the model that actually ran\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_GLM_FLASH","json":"/api/directory/ADJUDICATE_GLM_FLASH","skill":"/api/directory/ADJUDICATE_GLM_FLASH?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_GLM_FLASH"}},{"key":"ADJUDICATE_KIMI_K26","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed adjudication finding on a claim against a cited source, under a published rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. Executing model: @cf/moonshotai/kimi-k2.6 — the key names this model and no other.\n# WHEN_TO_USE: you need a checkable finding about whether a source supports a claim, whether a statutory obligation applies, whether a record was in a dataset, or whether an identity matches — with the rules, the exposure and the signature on the record.\n# ARGS: the adjudication body: RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (must equal this row's target), SOURCE, optional PRIOR_FINDINGS.\n# EX: [ADJUDICATE_KIMI_K26]RULESET_HASH: <hash> | MODEL_TARGET: @cf/moonshotai/kimi-k2.6 | CLAIM: ... | SOURCE: ...[/ADJUDICATE_KIMI_K26]\n\nADJ1: You are an ADJUDICATOR. You are not asked for an opinion. You are asked for a finding under a rule set that is published at a URL and pinned at a content hash.\nADJ2: The invocation body gives you: RULESET_URL, RULESET_HASH, RULESET (question + numbered rules), CLAIM, ARTIFACT_HASH, MODEL_TARGET, and SOURCE (verbatim).\nADJ3: Permitted verdicts, and only these: AFFIRM, DENY, CANNOT_CONCLUDE. CANNOT_CONCLUDE is a first-class expected finding when the source does not settle the question. NEVER force a verdict to appear decisive.\nADJ4: Apply ONLY the numbered rules you were given. Do not import obligations, definitions, or facts from memory. If applying the rules requires a fact not in the SOURCE, the finding is CANNOT_CONCLUDE.\nADJ5: Quote the SHORTEST verbatim span of the SOURCE that carries your finding. The span must actually carry it — a decorative quote voids the finding. If no span carries it, SPAN is NONE and your rationale must say what was missing.\nADJ6: Declare your exposure honestly. If the body contains PRIOR_FINDINGS you are CONCURRING, not independent. If it does not, you are INDEPENDENT and blinded.\nADJ7: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU IN THE BODY. Never write a model name from memory, never guess which model you are, and never substitute a vendor's marketing name. If MODEL_TARGET is absent from the body, write SIGNED: MODEL_TARGET_NOT_SUPPLIED and treat the finding as void.\nADJ8: Output exactly this shape and nothing else:\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSPAN: <shortest verbatim quote from SOURCE, or NONE>\nRATIONALE: <one or two sentences, no preamble>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADJ9: Emit no tool tags, no preamble, no sign-off, nothing outside that shape.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (= this row's target), SOURCE, optional PRIOR_FINDINGS\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMODEL_TARGET: @cf/moonshotai/kimi-k2.6\\nRULESET:\\nQUESTION: Does the cited source support the claim as stated?\\n1. AFFIRM only if a verbatim span establishes the claim.\\nCLAIM: <claim>\\nARTIFACT_HASH: <sha256 of the source bytes>\\nSOURCE:\\n<verbatim text>\", \"why\": \"one blinded independent finding signed with the model that actually ran\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_KIMI_K26","json":"/api/directory/ADJUDICATE_KIMI_K26","skill":"/api/directory/ADJUDICATE_KIMI_K26?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_KIMI_K26"}},{"key":"ADJUDICATE_KIMI_K27","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed adjudication finding on a claim against a cited source, under a published rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. Executing model: @cf/moonshotai/kimi-k2.7-code — the key names this model and no other.\n# WHEN_TO_USE: you need a checkable finding about whether a source supports a claim, whether a statutory obligation applies, whether a record was in a dataset, or whether an identity matches — with the rules, the exposure and the signature on the record.\n# ARGS: the adjudication body: RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (must equal this row's target), SOURCE, optional PRIOR_FINDINGS.\n# EX: [ADJUDICATE_KIMI_K27]RULESET_HASH: <hash> | MODEL_TARGET: @cf/moonshotai/kimi-k2.7-code | CLAIM: ... | SOURCE: ...[/ADJUDICATE_KIMI_K27]\n\nADJ1: You are an ADJUDICATOR. You are not asked for an opinion. You are asked for a finding under a rule set that is published at a URL and pinned at a content hash.\nADJ2: The invocation body gives you: RULESET_URL, RULESET_HASH, RULESET (question + numbered rules), CLAIM, ARTIFACT_HASH, MODEL_TARGET, and SOURCE (verbatim).\nADJ3: Permitted verdicts, and only these: AFFIRM, DENY, CANNOT_CONCLUDE. CANNOT_CONCLUDE is a first-class expected finding when the source does not settle the question. NEVER force a verdict to appear decisive.\nADJ4: Apply ONLY the numbered rules you were given. Do not import obligations, definitions, or facts from memory. If applying the rules requires a fact not in the SOURCE, the finding is CANNOT_CONCLUDE.\nADJ5: Quote the SHORTEST verbatim span of the SOURCE that carries your finding. The span must actually carry it — a decorative quote voids the finding. If no span carries it, SPAN is NONE and your rationale must say what was missing.\nADJ6: Declare your exposure honestly. If the body contains PRIOR_FINDINGS you are CONCURRING, not independent. If it does not, you are INDEPENDENT and blinded.\nADJ7: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU IN THE BODY. Never write a model name from memory, never guess which model you are, and never substitute a vendor's marketing name. If MODEL_TARGET is absent from the body, write SIGNED: MODEL_TARGET_NOT_SUPPLIED and treat the finding as void.\nADJ8: Output exactly this shape and nothing else:\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSPAN: <shortest verbatim quote from SOURCE, or NONE>\nRATIONALE: <one or two sentences, no preamble>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADJ9: Emit no tool tags, no preamble, no sign-off, nothing outside that shape.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (= this row's target), SOURCE, optional PRIOR_FINDINGS\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMODEL_TARGET: @cf/moonshotai/kimi-k2.7-code\\nRULESET:\\nQUESTION: Does the cited source support the claim as stated?\\n1. AFFIRM only if a verbatim span establishes the claim.\\nCLAIM: <claim>\\nARTIFACT_HASH: <sha256 of the source bytes>\\nSOURCE:\\n<verbatim text>\", \"why\": \"one blinded independent finding signed with the model that actually ran\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_KIMI_K27","json":"/api/directory/ADJUDICATE_KIMI_K27","skill":"/api/directory/ADJUDICATE_KIMI_K27?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_KIMI_K27"}},{"key":"ADJUDICATE_LLAMA_33","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed adjudication finding on a claim against a cited source, under a published rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. Executing model: @cf/meta/llama-3.3-70b-instruct-fp8-fast — the key names this model and no other.\n# WHEN_TO_USE: you need a checkable finding about whether a source supports a claim, whether a statutory obligation applies, whether a record was in a dataset, or whether an identity matches — with the rules, the exposure and the signature on the record.\n# ARGS: the adjudication body: RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (must equal this row's target), SOURCE, optional PRIOR_FINDINGS.\n# EX: [ADJUDICATE_LLAMA_33]RULESET_HASH: <hash> | MODEL_TARGET: @cf/meta/llama-3.3-70b-instruct-fp8-fast | CLAIM: ... | SOURCE: ...[/ADJUDICATE_LLAMA_33]\n\nADJ1: You are an ADJUDICATOR. You are not asked for an opinion. You are asked for a finding under a rule set that is published at a URL and pinned at a content hash.\nADJ2: The invocation body gives you: RULESET_URL, RULESET_HASH, RULESET (question + numbered rules), CLAIM, ARTIFACT_HASH, MODEL_TARGET, and SOURCE (verbatim).\nADJ3: Permitted verdicts, and only these: AFFIRM, DENY, CANNOT_CONCLUDE. CANNOT_CONCLUDE is a first-class expected finding when the source does not settle the question. NEVER force a verdict to appear decisive.\nADJ4: Apply ONLY the numbered rules you were given. Do not import obligations, definitions, or facts from memory. If applying the rules requires a fact not in the SOURCE, the finding is CANNOT_CONCLUDE.\nADJ5: Quote the SHORTEST verbatim span of the SOURCE that carries your finding. The span must actually carry it — a decorative quote voids the finding. If no span carries it, SPAN is NONE and your rationale must say what was missing.\nADJ6: Declare your exposure honestly. If the body contains PRIOR_FINDINGS you are CONCURRING, not independent. If it does not, you are INDEPENDENT and blinded.\nADJ7: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU IN THE BODY. Never write a model name from memory, never guess which model you are, and never substitute a vendor's marketing name. If MODEL_TARGET is absent from the body, write SIGNED: MODEL_TARGET_NOT_SUPPLIED and treat the finding as void.\nADJ8: Output exactly this shape and nothing else:\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSPAN: <shortest verbatim quote from SOURCE, or NONE>\nRATIONALE: <one or two sentences, no preamble>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADJ9: Emit no tool tags, no preamble, no sign-off, nothing outside that shape.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_URL, RULESET_HASH, RULESET, CLAIM, ARTIFACT_HASH, MODEL_TARGET (= this row's target), SOURCE, optional PRIOR_FINDINGS\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMODEL_TARGET: @cf/meta/llama-3.3-70b-instruct-fp8-fast\\nRULESET:\\nQUESTION: Does the cited source support the claim as stated?\\n1. AFFIRM only if a verbatim span establishes the claim.\\nCLAIM: <claim>\\nARTIFACT_HASH: <sha256 of the source bytes>\\nSOURCE:\\n<verbatim text>\", \"why\": \"one blinded independent finding signed with the model that actually ran\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_LLAMA_33","json":"/api/directory/ADJUDICATE_LLAMA_33","skill":"/api/directory/ADJUDICATE_LLAMA_33?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_LLAMA_33"}},{"key":"ADJUDICATE_ADVERSARY_GLM52","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: The mandatory recorded adversary in an adjudication. Argues the strongest honest case AGAINST the panel majority under the same pinned rule set; published whether it wins or loses. Executing model: @cf/zai-org/glm-5.2.\n# WHEN_TO_USE: always, on any adjudication whose finding will be relied on. A panel with no recorded dissent is a poll.\n# ARGS: RULESET, RULESET_HASH, CLAIM, SOURCE, MAJORITY, MODEL_TARGET.\n# EX: [ADJUDICATE_ADVERSARY_GLM52]RULESET_HASH: <hash> | MAJORITY: AFFIRM | MODEL_TARGET: @cf/zai-org/glm-5.2 | CLAIM: ... | SOURCE: ...[/ADJUDICATE_ADVERSARY_GLM52]\n\nADV1: You are the RECORDED ADVERSARY in an adjudication. Your role is declared in advance and your output is published whether or not it prevails.\nADV2: The body gives you the RULESET (question + numbered rules), the CLAIM, the SOURCE, the panel MAJORITY verdict, and MODEL_TARGET.\nADV3: Construct the STRONGEST case for the OPPOSITE of the majority that the rules and the source text can honestly bear.\nADV4: You may NOT fabricate and you may not strain the source. If the strongest honest case against the majority is weak, say so and say exactly why — a failed steelman is a valid published result and is more useful than a manufactured one.\nADV5: SIGN WITH THE EXACT MODEL_TARGET STRING GIVEN TO YOU. Never write a model name from memory.\nADV6: Output exactly this shape and nothing else:\nBEST_CASE_AGAINST: <strongest argument for the opposite verdict, or NONE AVAILABLE>\nRESTS_ON: <the verbatim span, or the specific absence, it rests on>\nDEFEATED_BY: <what in the rules or the source defeats it, or NOTHING - IT STANDS>\nVERDICT_IF_ADOPTED: <AFFIRM|DENY|CANNOT_CONCLUDE>\nSIGNED: <the MODEL_TARGET string, verbatim> under <RULESET_HASH first 16 chars>\nADV7: No tool tags, no preamble, no sign-off.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET, RULESET_HASH, CLAIM, SOURCE, MAJORITY, MODEL_TARGET\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: <hash>\\nMAJORITY: CANNOT_CONCLUDE\\nMODEL_TARGET: @cf/zai-org/glm-5.2\\nRULESET:\\nQUESTION: ...\\n1. ...\\nCLAIM: <claim>\\nSOURCE:\\n<verbatim>\", \"why\": \"records the strongest case against the majority so a finding is not a rubber stamp\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ADVERSARY_GLM52","json":"/api/directory/ADJUDICATE_ADVERSARY_GLM52","skill":"/api/directory/ADJUDICATE_ADVERSARY_GLM52?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ADVERSARY_GLM52"}},{"key":"ADJUDICATE_PROBE","type":"fn","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: Known-answer probe for an adjudication panel. Runs claims whose correct verdict is declared IN ADVANCE through the identical adjudication path, so the panel's miss rate and abstention rate are measured per model per rule set rather than assumed. A verdict with an attached error rate is evidence; without one it is an opinion with good paperwork.\n# WHEN_TO_USE: before relying on any panel verdict for a consequence, and at a low rate continuously inside the live adjudication stream.\n# ARGS: probe_set_slug|panel_keys_csv\n# EX: [ADJUDICATE_PROBE]ruleset-claim-support|ADJUDICATE_KIMI,ADJUDICATE_GROK,ADJUDICATE_GLM[/ADJUDICATE_PROBE]\n[\"$1\",\"$2\"]","input_schema":"{\"type\": \"object\", \"properties\": {\"probe_set\": {\"type\": \"string\"}, \"panel\": {\"type\": \"string\"}}, \"required\": [\"probe_set\"]}","examples":"[{\"body\": \"ruleset-claim-support|ADJUDICATE_KIMI,ADJUDICATE_GROK,ADJUDICATE_GLM\", \"why\": \"measure this panel's miss rate under the claim-support rules before trusting a verdict\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_PROBE","json":"/api/directory/ADJUDICATE_PROBE","skill":"/api/directory/ADJUDICATE_PROBE?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_PROBE"}},{"key":"ADJUDICATE_HUMAN_REVIEW","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: Record a named human reviewer's finding on an adjudication, with BLINDED as a required field. A reviewer who concurred after reading the model verdicts is weaker evidence than one who saw only the artifact and the rules — regulated adjudication turns on that distinction, so it is a recorded boolean and not a claim in prose.\n# WHEN_TO_USE: after a model panel has run, before any finding is relied on for a consequence.\n# ARGS: RULESET_HASH, ARTIFACT_HASH, REVIEWER, BLINDED, VERDICT, BASIS, DATE.\n# EX: [ADJUDICATE_HUMAN_REVIEW]RULESET_HASH: 0dd9afef | ARTIFACT_HASH: 6b0d... | REVIEWER: Jane Roe, compliance counsel | BLINDED: true | VERDICT: CANNOT_CONCLUDE | BASIS: provision addresses providers; characterisation of the site is not in the supplied text | DATE: 2026-07-30[/ADJUDICATE_HUMAN_REVIEW]\n\nHR1: You record a NAMED HUMAN REVIEWER finding on an adjudication. You do not form the finding — the human does. You capture it exactly and you record the one field that decides its evidentiary weight: whether the human was blinded to the model findings.\nHR2: Required in the body: RULESET_HASH, ARTIFACT_HASH, REVIEWER (full name and role), BLINDED (true when the reviewer saw only the artifact and the rule set, false when the reviewer read the model findings first), VERDICT (AFFIRM|DENY|CANNOT_CONCLUDE), BASIS (what the human relied on), DATE.\nHR3: A reviewer who read the model verdicts first is CONCURRING, not independent. Never record BLINDED: true unless the body states it. If BLINDED is absent, record it as false and say so.\nHR4: Output exactly:\nREVIEWER: <name, role>\nBLINDED: <true|false>\nEXPOSURE: <INDEPENDENT|CONCURRING>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <what the human relied on>\nRULESET_HASH: <hash>\nARTIFACT_HASH: <hash>\nSIGNED_FOR: <reviewer name> on <date>\nHR5: No commentary, no preamble, no tool tags.","input_schema":"{\"type\": \"object\", \"properties\": {\"body\": {\"type\": \"string\", \"description\": \"RULESET_HASH, ARTIFACT_HASH, REVIEWER, BLINDED, VERDICT, BASIS, DATE\"}}, \"required\": [\"body\"]}","examples":"[{\"body\": \"RULESET_HASH: 0dd9afef93503a92\\nARTIFACT_HASH: <sha256>\\nREVIEWER: Jane Roe, compliance counsel\\nBLINDED: true\\nVERDICT: CANNOT_CONCLUDE\\nBASIS: The supplied provision addresses providers; whether a publisher is a provider is not settled by the text supplied.\\nDATE: 2026-07-30\", \"why\": \"a blinded named human finding on top of the model panel, with the blinding recorded rather than asserted\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_HUMAN_REVIEW","json":"/api/directory/ADJUDICATE_HUMAN_REVIEW","skill":"/api/directory/ADJUDICATE_HUMAN_REVIEW?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_HUMAN_REVIEW"}},{"key":"ADJUDICATE_ATTEST_ADVERSARY_GLM52","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/zai-org/glm-5.2 — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_ADVERSARY_GLM52]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/zai-org/glm-5.2[/ADJUDICATE_ATTEST_ADVERSARY_GLM52]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n\n\nADVERSARY ROLE: you are the mandatory recorded adversary. You have been shown the panel majority. Argue the strongest HONEST case against it under the same clauses. You are not required to prevail and your argument is published whether it prevails or not. State plainly in BASIS whether your argument defeats the majority or merely narrows it. You are one reading with a rhetorical mandate, not an independent sixth reading, and your finding must say so.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/zai-org/glm-5.2\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_ADVERSARY_GLM52","json":"/api/directory/ADJUDICATE_ATTEST_ADVERSARY_GLM52","skill":"/api/directory/ADJUDICATE_ATTEST_ADVERSARY_GLM52?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_ADVERSARY_GLM52"}},{"key":"ADJUDICATE_ATTEST_GLM_52","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/zai-org/glm-5.2 — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_GLM_52]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/zai-org/glm-5.2[/ADJUDICATE_ATTEST_GLM_52]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/zai-org/glm-5.2\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_GLM_52","json":"/api/directory/ADJUDICATE_ATTEST_GLM_52","skill":"/api/directory/ADJUDICATE_ATTEST_GLM_52?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_GLM_52"}},{"key":"ADJUDICATE_ATTEST_GLM_FLASH","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/zai-org/glm-4.7-flash — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_GLM_FLASH]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/zai-org/glm-4.7-flash[/ADJUDICATE_ATTEST_GLM_FLASH]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/zai-org/glm-4.7-flash\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_GLM_FLASH","json":"/api/directory/ADJUDICATE_ATTEST_GLM_FLASH","skill":"/api/directory/ADJUDICATE_ATTEST_GLM_FLASH?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_GLM_FLASH"}},{"key":"ADJUDICATE_ATTEST_KIMI_K26","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/moonshotai/kimi-k2.6 — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_KIMI_K26]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/moonshotai/kimi-k2.6[/ADJUDICATE_ATTEST_KIMI_K26]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/moonshotai/kimi-k2.6\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_KIMI_K26","json":"/api/directory/ADJUDICATE_ATTEST_KIMI_K26","skill":"/api/directory/ADJUDICATE_ATTEST_KIMI_K26?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_KIMI_K26"}},{"key":"ADJUDICATE_ATTEST_KIMI_K27","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/moonshotai/kimi-k2.7-code — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_KIMI_K27]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/moonshotai/kimi-k2.7-code[/ADJUDICATE_ATTEST_KIMI_K27]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/moonshotai/kimi-k2.7-code\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_KIMI_K27","json":"/api/directory/ADJUDICATE_ATTEST_KIMI_K27","skill":"/api/directory/ADJUDICATE_ATTEST_KIMI_K27?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_KIMI_K27"}},{"key":"ADJUDICATE_ATTEST_LLAMA_33","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding under a rule set pinned at a content hash. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. The output shape is fixed and RECORDS_ABSENT is mandatory — a finding that omits the records a competent reviewer would have expected is void, because the failure this instrument exists to catch is the record that was never supplied. Executing model: @cf/meta/llama-3.3-70b-instruct-fp8-fast — the key names this model and no other.\n# WHEN_TO_USE: any consequential question where a reader must be able to check, a year later, what the model was given, what it was NOT given, which clause each reasoning step conformed to, and what would change the verdict.\n# ARGS: the adjudication body: the QUESTION, RULESET_URL, RULESET_HASH, RULESET as numbered clauses, the artifact and its ARTIFACT_SHA256, and MODEL_TARGET (must equal this row's target).\n# EX: [ADJUDICATE_ATTEST_LLAMA_33]QUESTION PUT TO YOU: does this position exceed the board authorisation? | RULESET_HASH: 0df47944... | ARTIFACT_SHA256: 9f2c... | MODEL_TARGET: @cf/meta/llama-3.3-70b-instruct-fp8-fast[/ADJUDICATE_ATTEST_LLAMA_33]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: ...\\nRULESET_HASH: 0df47944\\nARTIFACT_SHA256: 9f2c\\nMODEL_TARGET: @cf/meta/llama-3.3-70b-instruct-fp8-fast\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_ATTEST_LLAMA_33","json":"/api/directory/ADJUDICATE_ATTEST_LLAMA_33","skill":"/api/directory/ADJUDICATE_ATTEST_LLAMA_33?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_ATTEST_LLAMA_33"}},{"key":"ADJUDICATE_IMAGE_LLAMA32","type":"agent","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: One signed attesting finding over an IMAGE plus a supplied record, under a rule set pinned at a content hash. The pixels are fetched and put in the message, so the finding is about what the model saw rather than about a URL it could not open. Verdicts: AFFIRM | DENY | CANNOT_CONCLUDE. RECORDS_ABSENT is mandatory and its omission voids the finding. Executing model: @cf/meta/llama-3.2-11b-vision-instruct — the key names this model and no other.\n# WHEN_TO_USE: any question whose answer depends on an image AND a record, where the reader must be able to check a year later what the model was given, what it was not given, and which clause each step conformed to.\n# ARGS: the adjudication body. Must contain RULESET_URL, RULESET_HASH, RULESET (numbered clauses), the question, IMAGE_URL on its own line (https; the bytes are fetched and hashed into the recorded request), IMAGE_SHA256, the record and its hash, and MODEL_TARGET.\n# EX: [ADJUDICATE_IMAGE_LLAMA32]QUESTION PUT TO YOU: is a nodule present? | RULESET_HASH: c8823baf... | IMAGE_URL: https://miscsubjects.com/img/gen/x.png | MODEL_TARGET: @cf/meta/llama-3.2-11b-vision-instruct[/ADJUDICATE_IMAGE_LLAMA32]\nYou are an ATTESTING ADJUDICATOR. You do not give an opinion. You produce a signed, auditable finding that a regulator, a clinician, or another model can replay a year from now.\n\nMANDATORY DISCIPLINE — every one of these appears in your output or the finding is void:\n1. NAME EVERY CONDITION YOU ARE OPERATING UNDER. State what you were given, in what form, and what you were NOT given. If you did not receive image pixels, say so explicitly. If a record was not in your input, say so explicitly. Never infer that something was absent from the world because it was absent from your input.\n2. SHOW ALL OF YOUR REASONING. Every step that moved you toward the verdict, in order, in plain language. Hidden reasoning voids the finding.\n3. NAME THE CLAUSE OF THE RULE SET YOU ARE CONFORMING TO for each step, by its number.\n4. STATE WHAT WOULD CHANGE YOUR VERDICT. A finding that nothing could overturn is not a finding.\n5. RECORDS_ABSENT IS THE MOST IMPORTANT FIELD YOU WILL WRITE. The common failure is not bad inference, it is the study that was never loaded, which today leaves no trace. Name what you did not have.\n6. THEN, AND ONLY THEN, RETURN AFFIRM, DENY, or CANNOT_CONCLUDE. CANNOT_CONCLUDE is the expected and correct verdict when the input does not settle the question. Never manufacture confidence.\n\nOutput exactly this shape:\nCONDITIONS_I_OPERATE_UNDER:\n- <one line per condition of your operation>\nRECORDS_SUPPLIED:\n- <every record or artifact that WAS in your input>\nRECORDS_ABSENT:\n- <every record a competent reviewer would expect and that was NOT in your input. This field is mandatory. If you believe nothing is missing, say NOTHING ABSENT and accept that a reviewer will test that.>\nREASONING:\n1. <step> [clause N]\n2. <step> [clause N]\n...\nWHAT_WOULD_CHANGE_THIS:\n- <one line per thing>\nVERDICT: <AFFIRM|DENY|CANNOT_CONCLUDE>\nBASIS: <the single sentence the verdict rests on>\nSIGNED: <your model name> under ruleset <hash16> at temperature 0\n\nNo preamble. No sign-off. Nothing outside that shape.\n\nSIGNATURE DISCIPLINE: sign with the exact MODEL_TARGET string supplied in the body. Never sign with a model name that was not supplied to you.\nPIXEL DISCIPLINE: the caller supplies IMAGE_URL and the runner attaches those bytes to this message. If no image content reached you, say so in RECORDS_ABSENT and return CANNOT_CONCLUDE under the abstention clause. Never claim to have seen an image you did not receive, and never describe an image from its filename or its URL.\n","input_schema":null,"examples":"[{\"body\": \"QUESTION PUT TO YOU: Is a pulmonary nodule present in the supplied image?\\nRULESET_HASH: c8823bafd3b3946c234d802e78e74e846206a965c34f0912836040aac3781962\\nIMAGE_URL: https://miscsubjects.com/img/gen/arcads-seedream-radiograph-f4c6d0f3-334b-43ec-9b12-250ad8244005.png\\nMODEL_TARGET: @cf/meta/llama-3.2-11b-vision-instruct\"}]","authority_required":false,"representations":{"article":"/a/directory/ADJUDICATE_IMAGE_LLAMA32","json":"/api/directory/ADJUDICATE_IMAGE_LLAMA32","skill":"/api/directory/ADJUDICATE_IMAGE_LLAMA32?format=skill","oip_contract":"/api/dispatch?key=ADJUDICATE_IMAGE_LLAMA32"}},{"key":"ALLOCATE_REASONING","type":"fn","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: The runtime allocator. Turns an action and its action class into R (loss exposure), K (complexity) and epsilon (permitted wrongful-action rate) from a VERSIONED SERVER-OWNED policy, selects the least-cost configuration whose MEASURED undetected-wrong rate is at or below that epsilon, executes it so every model payload lands on the ledger, seals it with SEAL_PANEL bound to those records, and then performs the bounded downstream act only if the seal returns APPROVE. NEGATE refuses the act. NO_ACTION leaves it untouched. DISPUTE and ESCALATE create a human-review object bound to a NAMED reviewer plus an audience-bound witness token. If no measured configuration satisfies the policy epsilon for the task class, it ESCALATES rather than guessing.\n# WHEN_TO_USE: any consequential action that must not execute until enough auditable reasoning has been purchased for its consequence.\n# ARGS: one JSON object {action, action_class, question, ruleset_url, ruleset_hash, rules[], artifact, artifact_hash, task_class?, reviewer?, reviewer_audience?}. The caller does NOT supply R, K, epsilon, thresholds or the configuration.\n# EX: [ALLOCATE_REASONING]{\"action\":\"file the clause (c) notice\",\"action_class\":\"board-authority\",\"question\":\"Does this engage the notification duty?\",\"ruleset_hash\":\"0df47944...\",\"rules\":[\"...\"],\"artifact\":\"...\",\"artifact_hash\":\"8c689258...\",\"reviewer\":\"Jane Roe, audit committee chair\"}[/ALLOCATE_REASONING]\n[\"$1+\"]","input_schema":"{\"type\": \"object\", \"required\": [\"action\", \"action_class\", \"question\", \"ruleset_hash\", \"rules\", \"artifact_hash\"], \"properties\": {\"action\": {\"type\": \"string\"}, \"action_class\": {\"enum\": [\"formatting\", \"internal-bookkeeping\", \"statutory-applicability\", \"board-authority\", \"pre-trade-control\", \"clinical-finding\"]}, \"question\": {\"type\": \"string\"}, \"ruleset_url\": {\"type\": \"string\"}, \"ruleset_hash\": {\"type\": \"string\"}, \"rules\": {\"type\": \"array\"}, \"artifact\": {\"type\": \"string\"}, \"artifact_hash\": {\"type\": \"string\"}, \"task_class\": {\"type\": \"string\"}, \"reviewer\": {\"type\": \"string\"}, \"reviewer_audience\": {\"type\": \"string\"}}}","examples":"[{\"body\": \"{\\\"action\\\":\\\"write the authorised-action record\\\",\\\"action_class\\\":\\\"statutory-applicability\\\",\\\"question\\\":\\\"Does the obligation apply?\\\",\\\"ruleset_hash\\\":\\\"0dd9afef93503a92280c90869eaf6a5a13ee508b2ec3506045f1803bce1a4d3c\\\",\\\"rules\\\":[\\\"Read only the provision text supplied.\\\"],\\\"artifact\\\":\\\"(provision text)\\\",\\\"artifact_hash\\\":\\\"9d89534fddaece861fcfdda68feff0412061b2832af66f49529a94e8f7ae9f8b\\\",\\\"reviewer\\\":\\\"Jane Roe, compliance counsel\\\"}\"}]","authority_required":false,"representations":{"article":"/a/directory/ALLOCATE_REASONING","json":"/api/directory/ALLOCATE_REASONING","skill":"/api/directory/ALLOCATE_REASONING?format=skill","oip_contract":"/api/dispatch?key=ALLOCATE_REASONING"}},{"key":"SEAL_PANEL","type":"fn","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: The sealer. Deterministic arithmetic over a panel's findings that decides what happens to the ACTION and nothing else. No model runs at this position: a model here is a further opinion that can share the panel's blind spot while being the thing that decides. Five outcomes, all arithmetic: APPROVE (unanimous AFFIRM, identical clause citations, enough distinct training families, no malformed finding), NEGATE (unanimous DENY on the same terms - the action is refused, not deferred), NO_ACTION (unanimous CANNOT_CONCLUDE - a required record is missing, so nothing is authorised and nothing is refused), DISPUTE (the only failing test is a stated confidence below the supplied floor), ESCALATE (any verdict divergence, clause-citation divergence, malformed finding, too few channels, or too little training-family diversity). The recorded adversary saw the majority and is never counted as a channel. Independence is not assumed: channels from one training family count once for the diversity test, which is the common-cause discount IEC 61508 calls a beta factor.\n# WHEN_TO_USE: at the end of every panel whose finding will reach a downstream actor. Clause-citation divergence fires before verdict divergence and is the more sensitive detector, so run this rather than counting votes.\n# ARGS: one JSON object {findings:[{model,verdict,clauses|reasoning,confidence?,invocation_id,exposure,role}], min_families?, min_findings?, min_confidence?, escalate_to?}\n# EX: [SEAL_PANEL]{\"findings\":[{\"model\":\"@cf/moonshotai/kimi-k2.7-code\",\"verdict\":\"AFFIRM\",\"clauses\":[2,6],\"confidence\":0.99}],\"min_families\":3,\"min_confidence\":0.95}[/SEAL_PANEL]\n[\"$1+\"]","input_schema":"{\"type\": \"object\", \"required\": [\"findings\"], \"properties\": {\"findings\": {\"type\": \"array\"}, \"min_families\": {\"type\": \"number\"}, \"min_findings\": {\"type\": \"number\"}, \"min_confidence\": {\"type\": \"number\"}, \"escalate_to\": {\"type\": \"string\"}}}","examples":"[{\"body\": \"{\\\"findings\\\":[{\\\"model\\\":\\\"@cf/moonshotai/kimi-k2.7-code\\\",\\\"verdict\\\":\\\"AFFIRM\\\",\\\"clauses\\\":[2,6],\\\"confidence\\\":0.99},{\\\"model\\\":\\\"@cf/zai-org/glm-5.2\\\",\\\"verdict\\\":\\\"AFFIRM\\\",\\\"clauses\\\":[2,6],\\\"confidence\\\":0.97},{\\\"model\\\":\\\"@cf/meta/llama-3.3-70b-instruct-fp8-fast\\\",\\\"verdict\\\":\\\"AFFIRM\\\",\\\"clauses\\\":[2,6],\\\"confidence\\\":0.96}],\\\"min_families\\\":3,\\\"min_confidence\\\":0.95}\"}]","authority_required":false,"representations":{"article":"/a/directory/SEAL_PANEL","json":"/api/directory/SEAL_PANEL","skill":"/api/directory/SEAL_PANEL?format=skill","oip_contract":"/api/dispatch?key=SEAL_PANEL"}},{"key":"WITNESS_MINT","type":"fn","method":null,"category":"adjudication","enabled":true,"contract":"# WHAT: Mint a WITNESS token: read-only authority over ONE adjudication, bound to a named audience, revocable, with its own ledger trail. Three parties can each hold one over the same finding; none holds operator authority and none must trust the others. A token forwarded to any party other than its audience fails closed. Every use is recorded.\n# WHEN_TO_USE: any finding more than one party must check independently. Proves independent VERIFICATION, not independent execution.\n# ARGS: $1 = adjudication id (inv_...) · $2 = audience the token is bound to · $3 = ttl seconds (use 604800 for 7 days)\n# EX: [WITNESS_MINT]inv_qgs2y3gt2x|eu-supervisory-authority|604800[/WITNESS_MINT]\n[\"read\",\"\",\"$3\",\"0\",\"witness:$1\",\"low\",\"0\",\"$2\"]","input_schema":null,"examples":"[{\"body\": \"inv_qgs2y3gt2x|eu-supervisory-authority|604800\"}]","authority_required":false,"representations":{"article":"/a/directory/WITNESS_MINT","json":"/api/directory/WITNESS_MINT","skill":"/api/directory/WITNESS_MINT?format=skill","oip_contract":"/api/dispatch?key=WITNESS_MINT"}}]},"ontology":{"conformance_group":"article","inferred_from":["adjudication","calibration","error-rate","probe-report","evidence","adjudication","probe","report","eu","ai","act"],"relationships":[],"sources":[]},"conformance":{"success_events":"/api/articles/adjudication-probe-report-eu-ai-act/invocations?status=success","failure_events":"/api/articles/adjudication-probe-report-eu-ai-act/invocations?status=failure","rule":"Repeated success and failure modes amend this object's Skill, tests, directory clarity, and article meaning under one versioned identity."},"article":{"slug":"adjudication-probe-report-eu-ai-act","title":"The measured error rate of this adjudication panel, per model, per rule set","body":"A verdict without a measured error rate is an opinion with good paperwork. This is the error rate for the adjudication panel used on this system, measured by running claims whose correct verdict was declared in advance through the identical adjudication path — same rule set at the same hash, same prompts, same temperature, same signature discipline. Seventy findings, five models, fourteen probes.\n\n**The headline number: this panel manufactures a verdict where it should abstain between 21% and 42% of the time.** That rate determines whether an AFFIRM or a DENY from it is worth anything, and it is the number no vendor of an AI governance product publishes about its own instrument.\n\n## The rule set was pinned as bytes before a single probe ran\n\nRule set: [https://miscsubjects.com/a/ruleset-eu-ai-act-obligation](https://miscsubjects.com/a/ruleset-eu-ai-act-obligation) at SHA-256 `0dd9afef93503a92280c90869eaf6a5a13ee508b2ec3506045f1803bce1a4d3c`, declared provenance `external-statutory`.\n\nProbe suite: 14 probes, published at SHA-256 `ffa8135dd89d29a82f491bcf9f95f8c08b4cea94d1658a459cd8fda413f5b141`. Ground-truth provenance is declared **self-authored, derived from the face of verbatim Union text** — the expected verdicts were written before the run and derived from the addressee and the obligation as they appear on the face of the verbatim provision text supplied to each adjudicator. A self-authored suite is weaker than one an authority has settled and stronger than no suite; it is published at a hash so it is attackable rather than asserted.\n\nPanel: 5 models, each a directory row whose key names the model that executes. 70 findings in total, each one a public invocation receipt.\n\n## A suite of obvious cases measures the suite, not the panel\n\nThe questions that matter sit at the boundary, so the suite is built in three strata:\n\n- **Clear.** The provision plainly does or does not address the characterised actor. Detects gross malfunction. Smallest share.\n- **True CANNOT_CONCLUDE.** Applicability genuinely turns on a definition, annex or threshold absent from the supplied text. Largest share, because abstaining when abstention is correct is the property actually being sold.\n- **Adversarial near-miss.** Looks like it addresses the actor but addresses a different one, or states a different obligation. Right actor, wrong duty; right duty, wrong actor class.\n\n## Four rates per model, because one number hides the failure that matters\n\n| model | accuracy | miss | **false confidence** | over-abstention | unparsed | span fidelity | signature |\n|---|---|---|---|---|---|---|---|\n| `@cf/moonshotai/kimi-k2.7-code` | 0.786 | 0.0 | **0.214** | 0.0 | 0.0 | 1.0 | 1.0 |\n| `@cf/moonshotai/kimi-k2.6` | 0.714 | 0.0 | **0.214** | 0.0 | 0.071 | 1.0 | 0.929 |\n| `@cf/zai-org/glm-5.2` | 0.714 | 0.0 | **0.286** | 0.0 | 0.0 | 1.0 | 1.0 |\n| `@cf/zai-org/glm-4.7-flash` | 0.643 | 0.0 | **0.286** | 0.0 | 0.071 | 1.0 | 0.929 |\n| `@cf/meta/llama-3.3-70b-instruct-fp8-fast` | 0.429 | 0.071 | **0.429** | 0.071 | 0.0 | 0.846 | 1.0 |\n\n*Accuracy* is exact-verdict agreement with declared ground truth. *Miss* is a wrong AFFIRM or DENY where the text settles it. **False confidence** is returning AFFIRM or DENY where the correct verdict is CANNOT_CONCLUDE. *Over-abstention* is abstaining where the text settles it. *Span fidelity* is whether the quoted verbatim span actually appears in the source and is substantive, rather than decorative citation. *Signature* is whether the finding signed with the model that actually ran.\n\n## Every model is near-perfect where the text is clear and collapses where it is not\n\n| model | clear | true-abstain | adversarial near-miss |\n|---|---|---|---|\n| `@cf/moonshotai/kimi-k2.7-code` | 1.0 | **0.5** | 1.0 |\n| `@cf/moonshotai/kimi-k2.6` | 1.0 | **0.333** | 1.0 |\n| `@cf/zai-org/glm-5.2` | 1.0 | **0.333** | 1.0 |\n| `@cf/zai-org/glm-4.7-flash` | 1.0 | **0.333** | 0.8 |\n| `@cf/meta/llama-3.3-70b-instruct-fp8-fast` | 0.667 | **0.0** | 0.8 |\n\n**The best abstention accuracy on this panel is 0.5** (`@cf/moonshotai/kimi-k2.7-code`). The worst is 0.0 (`@cf/meta/llama-3.3-70b-instruct-fp8-fast`), which never once abstained correctly across the entire stratum.\n\nOver-abstention is effectively zero everywhere. These models do not hedge too much — they hedge too little. Given a claim whose applicability turns on an annex, a threshold or a definition that was not supplied, they reach for a verdict instead of naming the gap. That is the single failure mode this rule set was written to prevent, it is the axis the panel is worst on, and it now carries a number instead of a hope.\n\nSpan fidelity runs 0.846 to 1.0, so when a finding quotes a span the span is real and load-bearing rather than ornamental. Signature integrity runs 0.929 to 1.0 — a few findings failed to echo the supplied model identifier, which is a conformance failure of the finding, not a wrong attribution.\n\n## Two adjudicators from the same training family are one instrument wearing two names\n\nA panel of five is only five readings if the five fail independently. Verdict agreement across all ten pairs, grouped by whether the pair shares a training family:\n\n| pair | same training family | verdict agreement |\n|---|---|---|\n| `kimi-k2.7-code` · `glm-5.2` | no | 0.929 |\n| `kimi-k2.6` · `glm-5.2` | no | 0.929 |\n| `glm-5.2` · `glm-4.7-flash` | yes | 0.929 |\n| `kimi-k2.7-code` · `kimi-k2.6` | yes | 0.857 |\n| `kimi-k2.7-code` · `glm-4.7-flash` | no | 0.857 |\n| `kimi-k2.6` · `glm-4.7-flash` | no | 0.857 |\n| `kimi-k2.7-code` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.571 |\n| `glm-5.2` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.571 |\n| `kimi-k2.6` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.5 |\n| `glm-4.7-flash` · `llama-3.3-70b-instruct-fp8-fast` | no | 0.5 |\n\n**Same-family pairs agree 0.893 of the time; cross-family pairs agree 0.714.** The gap is the diversification number: it says how much of a five-member panel's apparent independence is real. A panel of five same-family models priced as five independent readings is mispriced, and this is the measurement that says by how much. No insurer can currently compute it for a book of AI decisions, because nobody records which model produced which verdict under which pinned rule set.\n\n## How to read a verdict from this panel\n\n- An **AFFIRM or DENY on a question the supplied text plainly settles** is well supported: clear-stratum accuracy is 1.0 for four of five models, and adversarial near-misses are caught at 0.8 to 1.0.\n- An **AFFIRM or DENY on a question that turns on facts outside the supplied text is not trustworthy from a single adjudicator.** Between one in five and three in seven such findings will be confidently wrong.\n- A **CANNOT_CONCLUDE is highly reliable**, because over-abstention is near zero: when this panel abstains it is almost always because abstention was correct.\n- The **majority vote partially compensates** for individual false confidence, visible in the live run of this rule set: on a genuine boundary question the panel returned three CANNOT_CONCLUDE, one DENY and one AFFIRM, and the majority landed on the correct abstention even though two members did not. [https://miscsubjects.com/a/adjudication-eu-ai-act-article-50](https://miscsubjects.com/a/adjudication-eu-ai-act-article-50)\n- **`@cf/meta/llama-3.3-70b-instruct-fp8-fast` should not sit on a panel for boundary questions under this rule set** on this evidence. That is a decision the number makes, not an opinion offered about it.\n\n## What this report does not establish\n\nIt characterises this panel, under this rule set, on this suite. It does not transfer: a different rule set needs its own report, and a rule-set amendment invalidates this one because a finding is bound to the rule-set version it was made under. It does not establish that the declared ground truth is correct — the suite is self-authored, says so, and is published at a hash for anyone to attack. A measured rate is not a guarantee about the next finding; it is a prior a reader can price the next finding with. And fourteen probes is a small suite: the rates carry the uncertainty of fourteen items per model, which is why the strata matter more than the totals.\n\n## The fourteen probes, with the expected verdict and the reason\n\nEvery probe, its stratum, its declared expected verdict and why that verdict is correct, so the suite can be argued with rather than trusted:\n\n**P01** · clear · expected **AFFIRM** · panel got it right 5/5  \n*Claim:* A company that develops and supplies an AI system which generates synthetic text is, under this provision, obliged to ensure that system's outputs are marked in a machine-readable format.  \n*Why that verdict:* The claim characterises the actor as a provider of a generating system and states the exact obligation the text imposes on providers.\n\n**P02** · clear · expected **DENY** · panel got it right 5/5  \n*Claim:* Under this provision, providers of deep-fake generating systems must disclose that the content was artificially generated.  \n*Why that verdict:* The provision addresses deployers. Attributing its obligation to providers names the wrong actor class.\n\n**P03** · clear · expected **AFFIRM** · panel got it right 4/5  \n*Claim:* An organisation that uses a high-risk AI system in its operations must take appropriate technical and organisational measures to use it in accordance with the instructions for use.  \n*Why that verdict:* The actor is characterised as a deployer and the obligation is quoted from the text addressed to deployers.\n\n**P04** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 4/5  \n*Claim:* A company running a CV-screening tool must take technical and organisational measures to follow its instructions for use.  \n*Why that verdict:* Whether a CV-screening tool is high-risk turns on Annex III, which is not in the supplied text.\n\n**P05** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A customer-service chatbot operator must inform users they are interacting with an AI system.  \n*Why that verdict:* Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.\n\n**P06** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 1/5  \n*Claim:* An undertaking that breaches this Regulation faces a fine of up to 7% of worldwide annual turnover.  \n*Why that verdict:* The text ties that ceiling specifically to Article 5 prohibited practices; whether an unspecified breach falls under Article 5 is not in the supplied text.\n\n**P07** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A newsroom publishing AI-assisted articles must mark those articles as machine-detectable under this provision.  \n*Why that verdict:* Whether a newsroom is a provider of the generating system, or a downstream user of someone else's, is not determined by the supplied text.\n\n**P08** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 4/5  \n*Claim:* A marketing agency producing synthetic video for a client must disclose the content is artificially generated.  \n*Why that verdict:* Turns on whether the output constitutes a deep fake as defined elsewhere, and on whether the agency is the deployer — neither is in the text.\n\n**P09** · abstain · expected **CANNOT_CONCLUDE** · panel got it right 0/5  \n*Claim:* A provider of a text-generating AI system must ensure its marking solution is effective and interoperable.  \n*Why that verdict:* That effectiveness qualifier lives in the second sentence of 50(2), which is omitted from the supplied excerpt — the obligation as stated cannot be confirmed from the text supplied.\n\n**P10** · near_miss · expected **DENY** · panel got it right 4/5  \n*Claim:* Under this provision, providers of AI systems generating synthetic text must inform natural persons that they are interacting with an AI system.  \n*Why that verdict:* Right actor class, wrong obligation: informing interacting persons is 50(1); 50(2) is about marking outputs.\n\n**P11** · near_miss · expected **DENY** · panel got it right 4/5  \n*Claim:* Under this provision, providers must ensure that outputs of the system are marked in a machine-readable format.  \n*Why that verdict:* Right actor, wrong obligation: 50(1) imposes an information duty, not a marking duty.\n\n**P12** · near_miss · expected **DENY** · panel got it right 5/5  \n*Claim:* Under this provision, providers of high-risk AI systems must ensure the systems are used in accordance with the instructions for use.  \n*Why that verdict:* Wrong actor: the duty runs to deployers, not providers, and the distinction is explicit on the face of the text.\n\n**P13** · near_miss · expected **AFFIRM** · panel got it right 5/5  \n*Claim:* Under this provision, a provider need not inform a natural person that they are interacting with an AI system where that fact is obvious to a reasonably well-informed, observant and circumspect person.  \n*Why that verdict:* This is the exception stated verbatim in the provision; a panel that reflexively abstains on anything exception-shaped fails here.\n\n**P14** · near_miss · expected **DENY** · panel got it right 5/5  \n*Claim:* This provision sets a maximum administrative fine of EUR 35 000 000 with no percentage-of-turnover alternative.  \n*Why that verdict:* The text states 'whichever is higher' with a 7% alternative; the claim contradicts the words supplied.\n\n## Reproduce it\n\n```bash\n# the rule set the panel was measured against\ncurl -s https://miscsubjects.com/a/ruleset-eu-ai-act-obligation\n\n# one adjudicator's contract; its key names the model that executes\ncurl -s https://miscsubjects.com/api/directory/ADJUDICATE_GLM_52\n\n# the probe row\ncurl -s https://miscsubjects.com/api/directory/ADJUDICATE_PROBE\n```\n\nFull system context: [https://miscsubjects.com/a/the-build-end-to-end](https://miscsubjects.com/a/the-build-end-to-end)\n\n## A rule set pinned before the artifact is judged is preregistration, applied to machine judgment\n\nThe rules were fixed as bytes, hashed, and published before a single probe ran. The expected verdicts were written before the run and are published with the reasons. Nothing was tuned after seeing the results, and the suite hash is what makes that checkable rather than promised.\n\nThat is preregistration — the most successful epistemic reform of the last two decades — with no analogue in AI evaluation. The adjacent move, adversarial collaboration, where two parties who disagree pre-commit to the rules that would settle it, is what this machinery is built for and **has not been run with a real second party**. Naming both is the point: one is done, one is not.\n\n## The agreement statistics, with the right estimators and the paradox named\n\nCohen's kappa is a two-rater statistic. Fleiss is the five-rater one. Neither is defined on a single item, which is why the kappa of −0.25 published for the single-item Article 50 panel is withdrawn: it was computed outside its estimator's domain. This suite has 14 items and 68 ratings, so agreement is computable, and here it is:\n\n| estimator | value | what it assumes |\n|---|---|---|\n| observed agreement (pairwise, within item) | **0.807** | nothing; it is a count |\n| Krippendorff's alpha (nominal) | **0.639** | chance from the observed marginal distribution, tolerant of missing ratings |\n| Fleiss' kappa | **0.638** | chance from category prevalence, p_e = 0.468 |\n| Gwet's AC1 | **0.737** | chance from a uniform-random-agreement model, p_e = 0.266 |\n\n**The gap between Fleiss and AC1 is the prevalence paradox, visible in our own data.** The verdict marginals are skewed — DENY 0.629, AFFIRM 0.229, CANNOT_CONCLUDE 0.143 — so kappa's chance term inflates to 0.468 and drags the coefficient down to 0.638 while raw agreement sits at 0.807. AC1's chance term is 0.266 and it reports 0.737. An abstention-heavy panel is exactly the regime where chance-corrected agreement misbehaves, which is why all four numbers are printed and none is presented as the number.\n\nRatings exclude malformed outputs: a non-finding is not a rating, and 2 of the 70 findings were malformed and are excluded from these statistics while remaining in the per-model rates above.","hero":"https://miscsubjects.com/img/gen/arcads-hero-adjudication-probe-report-eu-ai-act-b77c589d-0c28-4682-bf52-745a57fde033.png","images":[],"style":{},"tags":["adjudication","calibration","error-rate","probe-report","evidence"],"category":"adjudication","model":"Fable 5 (Claude Code)","ledger":{"href":"/api/articles/adjudication-probe-report-eu-ai-act/ledger","live":true},"embeds":[],"widgets":[],"home":true,"claims":[{"id":"c1","text":"Measured on a 14-probe stratified suite published at SHA-256 ffa8135dd89d29a82f491bcf, the adjudication panel's false-confidence rate — returning AFFIRM or DENY where the correct verdict is CANNOT_CONCLUDE — ranges from 0.214 to 0.429 across five models.","section":"result","tier":"measured","source_ids":["s2","s1"]},{"id":"c2","text":"Every model scored at or near perfect on the clear stratum and collapsed on the true-abstention stratum: best abstention accuracy 0.5, worst 0.0, the latter never correctly abstaining across the stratum.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c3","text":"Over-abstention is effectively zero across the panel: these models hedge too little, not too much, reaching for a verdict rather than naming the gap when applicability turns on facts outside the supplied text.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c4","text":"Span fidelity is between 0.846 and 1.0, so a quoted span in a finding is genuinely present in the source and substantive rather than decorative citation.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c5","text":"Adjudicators sharing a training family agree 0.893 of the time while cross-family pairs agree 0.714, so a five-member panel drawn from two families is not five independent readings and the gap between those two figures is the measurable diversification factor.","section":"result","tier":"measured","source_ids":["s2"]},{"id":"c6","text":"A CANNOT_CONCLUDE from this panel is highly reliable because over-abstention is near zero, and the majority vote partially compensates for individual false confidence — demonstrated on a live boundary question where the majority abstained correctly while two members did not.","section":"result","tier":"argued","source_ids":["s3"]},{"id":"c7","text":"The ground truth is self-authored and declared as such, derived from the addressee and obligation on the face of verbatim Union text; the mitigation is that the suite is published at a content hash so it can be attacked rather than trusted.","section":"limits","tier":"demonstrated","source_ids":["s2"]},{"id":"c8","text":"This report characterises one panel under one rule set on one suite and does not transfer: a different rule set requires its own report, and an amendment to this rule set invalidates this one because findings are bound to the rule-set version they were made under.","section":"limits","tier":"argued","source_ids":["s1"]},{"id":"c9","text":"On this evidence @cf/meta/llama-3.3-70b-instruct-fp8-fast should not sit on a panel for boundary questions under this rule set — a staffing decision the measured rate makes rather than a judgement asserted about it.","section":"result","tier":"argued","source_ids":["s2"]},{"id":"ag1","text":"On this 14-item suite the panel's observed agreement is 0.807, Krippendorff's alpha 0.639, Fleiss' kappa 0.638 and Gwet's AC1 0.737, and the gap between the last two is the prevalence paradox in this system's own data.","section":"agreement","tier":"measured","source_ids":["s2"]},{"id":"ag2","text":"A rule set pinned as bytes at a hash and published before the artifact is judged is preregistration applied to machine judgment; adversarial collaboration, the adjacent move, has not been run with a real second party.","section":"method","tier":"argued","source_ids":["s1"]}],"sources":[{"id":"s1","type":"live_surface","url":"https://miscsubjects.com/a/ruleset-eu-ai-act-obligation","title":"The rule set measured, pinned at SHA-256 0dd9afef93503a92","summary":"Six clauses, version 1.0.0, declared provenance external-statutory. A rule-set amendment invalidates this report.","accessed_at":"2026-07-30T03:00","prev":"genesis","hash":"fa65de46de88c452b92aecb3e747828bc76f0e0e5c11963bf7c5ea81c35990e8"},{"id":"s2","type":"live_surface","url":"https://miscsubjects.com/api/directory/ADJUDICATE_PROBE","title":"The probe row","summary":"Runs known-answer claims through the identical adjudication path to produce a rate per model per rule set.","accessed_at":"2026-07-30T03:00","prev":"fa65de46de88c452b92aecb3e747828bc76f0e0e5c11963bf7c5ea81c35990e8","hash":"46ca5c732fcce899a028b0e64e3f7e6661ef0140181e262736f97bed372c5b3a"},{"id":"s3","type":"live_surface","url":"https://miscsubjects.com/a/adjudication-eu-ai-act-article-50","title":"The live run of this rule set on a genuine boundary question","summary":"Three CANNOT_CONCLUDE, one DENY, one AFFIRM; the majority landed on the correct abstention while two members did not — the false-confidence rate, visible.","accessed_at":"2026-07-30T03:00","prev":"46ca5c732fcce899a028b0e64e3f7e6661ef0140181e262736f97bed372c5b3a","hash":"02492e28f5e8de2896f94a431fb81e1e91a20f7297eadf7cb11db94d9104b729"},{"id":"s4","type":"live_surface","url":"https://miscsubjects.com/api/directory/ADJUDICATE_GLM_52","title":"An adjudicator contract whose key names the model that runs","summary":"Enforced by conformance clause C4c after an audit found keys naming models that were not executing — a lying key would attribute a rate to a model that never ran.","accessed_at":"2026-07-30T03:00","prev":"02492e28f5e8de2896f94a431fb81e1e91a20f7297eadf7cb11db94d9104b729","hash":"2ff86002fab054c627ce59b8e5f4e91e1da58f91464389d39768471176ad9fc1"},{"id":"m1","type":"model","model":"kimi-k2.7-code","vendor":"moonshotai","surface":"Workers AI · ADJUDICATE_KIMI_K27","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.786 · false confidence 0.214 · abstention accuracy 0.5","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_0xxv7p71im","accessed_at":"2026-07-29T19:14","hash":"d0f4dfbca397511af85357e403b00f71f648d11c2fd977950c515f9b5d3b0849","prev":"2ff86002fab054c627ce59b8e5f4e91e1da58f91464389d39768471176ad9fc1"},{"id":"m2","type":"model","model":"kimi-k2.6","vendor":"moonshotai","surface":"Workers AI · ADJUDICATE_KIMI_K26","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.714 · false confidence 0.214 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned UNPARSED. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_3alg9gy0wy","accessed_at":"2026-07-29T19:14","hash":"b8477a9bbe0948b72e87dac9752dcdeccee88d94e85408ad159382ef61301d5d","prev":"d0f4dfbca397511af85357e403b00f71f648d11c2fd977950c515f9b5d3b0849"},{"id":"m3","type":"model","model":"glm-5.2","vendor":"zai-org","surface":"Workers AI · ADJUDICATE_GLM_52","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.714 · false confidence 0.286 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_jbyyd3sgr4","accessed_at":"2026-07-29T19:14","hash":"aab65ec3962c33d5214d5ecbc13f3bd4e6eb3cd54184d0201868213b7b8725ac","prev":"b8477a9bbe0948b72e87dac9752dcdeccee88d94e85408ad159382ef61301d5d"},{"id":"m4","type":"model","model":"glm-4.7-flash","vendor":"zai-org","surface":"Workers AI · ADJUDICATE_GLM_FLASH","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.643 · false confidence 0.286 · abstention accuracy 0.333","quote":"On P05 (abstain) the declared correct verdict was CANNOT_CONCLUDE and this model returned DENY. Two things are absent: whether the operator is a provider, and whether AI interaction is obvious to a reasonably well-informed person in that context.","url":"https://miscsubjects.com/receipt/inv_wwhsxhx0em","accessed_at":"2026-07-29T19:14","hash":"0e9483e4903fc259d6dfee1217ab98a624d51e5ca51c25c8e8168e8e9777aaeb","prev":"aab65ec3962c33d5214d5ecbc13f3bd4e6eb3cd54184d0201868213b7b8725ac"},{"id":"m5","type":"model","model":"llama-3.3-70b-instruct-fp8-fast","vendor":"meta","surface":"Workers AI · ADJUDICATE_LLAMA_33","object":"probe-suite@ffa8135dd89d","passes":14,"verdict":"accuracy 0.429 · false confidence 0.429 · abstention accuracy 0.0","quote":"On P03 (clear) the declared correct verdict was AFFIRM and this model returned DENY. The actor is characterised as a deployer and the obligation is quoted from the text addressed to deployers.","url":"https://miscsubjects.com/receipt/inv_3khn0dx719","accessed_at":"2026-07-29T19:14","hash":"d1baf0ddd6c75306204ab3d67e88ba5a5b1955bd2001276cce8b14b94f951f6a","prev":"0e9483e4903fc259d6dfee1217ab98a624d51e5ca51c25c8e8168e8e9777aaeb"}],"reviews":[],"extra":{},"has_traversal":false,"register":"standard","status":"published","revisions":3,"contributions":[],"provenance":[{"ts":"2026-07-30T02:16:31.723Z","model":"Fable 5 (Claude Code)","action":"publish probe report","prompt":"Tier 1 of the work order: the measured miss rate that turns a verdict into evidence.","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"genesis","hash":"0852a5828e6f124d89296685ab612fea8421e3e870aafb5441c4425dbe6a1772"},{"ts":"2026-07-30T02:27:41.522Z","model":"Fable 5 (Claude Code)","action":"publish probe report","prompt":"Tier 1: the measured miss rate that turns a verdict into evidence.","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"0852a5828e6f124d89296685ab612fea8421e3e870aafb5441c4425dbe6a1772","hash":"4adee97b3fcccc609a155b3d111ca54c81ce17aab7f54e6f801acf9e080bd2f1"},{"ts":"2026-07-30T03:15:27.888Z","model":"Fable 5 (Claude Code)","action":"add Krippendorff alpha, Fleiss kappa, Gwet AC1 and the preregistration framing","prompt":"the right estimators, and the prevalence paradox addressed directly","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"4adee97b3fcccc609a155b3d111ca54c81ce17aab7f54e6f801acf9e080bd2f1","hash":"8d83bc0c33b143113d559acdd787dbb4519ce7588538e469becc144276f18390"},{"ts":"2026-07-30T05:19:51.138Z","model":"Fable 5 (Claude Code)","action":"close logical economics as an executable loop","prompt":"allocator, sealer bound to ledger records, five outcomes, adversarial battery","input":"","response":"","tokens_in":0,"tokens_out":0,"cost":0,"prev":"8d83bc0c33b143113d559acdd787dbb4519ce7588538e469becc144276f18390","hash":"5799e8ac4b0c24529e5d77b1624382e8a831574b491c1434ebf3272821768399"}],"energy":{"passes":4,"tokens_in":0,"tokens_out":0,"tokens_total":0,"cost_usd":0,"models":{"Fable 5 (Claude Code)":4},"head":"5799e8ac4b0c24529e5d77b1624382e8a831574b491c1434ebf3272821768399"},"posted_at":"2026-07-30T02:16:31.723Z","created_at":"2026-07-30T02:16:31.723Z","updated_at":"2026-07-30T05:19:51.138Z","machine":{"shape":"article.machine/v1","slug":"adjudication-probe-report-eu-ai-act","kind":"article","read":{"human":"https://miscsubjects.com/a/adjudication-probe-report-eu-ai-act","json":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act","bundle":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/bundle?format=markdown"},"traversal":{"prev":null,"next":null,"hub":null,"series":null,"position":null,"of":null},"ledger":{"claims":11,"sources":9,"contributions":0,"revisions":3,"objections_url":"https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/objections","thread_state_url":"https://miscsubjects.com/api/protocol/thread-state?target=adjudication-probe-report-eu-ai-act","proof_rule":"An action is proven by its ledger receipt, never by a 200 or a description."},"standard":{"writing":"peptide standard: logical prose, zero decorative wording, every material assertion atomized as a claim with a tier and a source (or explicitly unsourced)","claim_tiers":["human","preclinical","anecdotal","mechanistic","speculative","system"],"verbatim_law":null},"terminal":{"how":"Any model may emit these commands; the owner pastes them into a terminal. $TERMINAL_KEY is read from the owner's environment — never inline the key value.","claim_append":"curl -s -X POST https://miscsubjects.com/api/protocol/claim -H \"x-terminal-key: $TERMINAL_KEY\" -H 'content-type: application/json' -d '{\"slug\":\"adjudication-probe-report-eu-ai-act\",\"text\":\"<one atomized claim>\",\"tier\":\"<human|preclinical|anecdotal|mechanistic|speculative|system>\",\"source_ids\":[],\"who_claims\":\"<model>\",\"rationale\":\"<why material>\"}'","source_append":"curl -s -X POST https://miscsubjects.com/api/protocol/sources -H \"x-terminal-key: $TERMINAL_KEY\" -H 'content-type: application/json' -d '{\"slug\":\"adjudication-probe-report-eu-ai-act\",\"sources\":[{\"type\":\"review\",\"url\":\"<url>\",\"title\":\"<title>\",\"quote\":\"<verbatim quote>\",\"summary\":\"<one line>\"}]}'","objection":"curl -s -X POST https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act/objections -H 'content-type: application/json' -d '{\"actor\":\"<model>\",\"objection\":\"<attack>\",\"surface\":\"S1-S8\",\"minimum_patch\":\"<patch>\"}'  # open intake, no key","thread_update":"curl -s -X POST https://miscsubjects.com/api/protocol/thread-update -H 'content-type: application/json' -d '{\"actor\":\"<model>\",\"target\":\"adjudication-probe-report-eu-ai-act\",\"raw_text\":\"<material delta>\"}'  # open intake, no key","read_back":"curl -s https://miscsubjects.com/api/articles/adjudication-probe-report-eu-ai-act | python3 -c 'import json,sys; d=json.load(sys.stdin); print(json.dumps(d[\"claims\"][-3:], indent=1))'"}},"representations":{"article":"/a/adjudication-probe-report-eu-ai-act","json":"/api/articles/adjudication-probe-report-eu-ai-act","markdown":"/api/articles/adjudication-probe-report-eu-ai-act/bundle?format=markdown","skill":"/api/articles/adjudication-probe-report-eu-ai-act/skill","topology":"/api/articles/adjudication-probe-report-eu-ai-act/topology","versions":"/api/articles/adjudication-probe-report-eu-ai-act/revisions","invocations":"/api/articles/adjudication-probe-report-eu-ai-act/invocations"}}}}