Propose testable catalyst hypotheses absent from the OCx25 dataset

Checks run in this order. Open a step to inspect its rules, checker or review rubric.

1. Result document structureAutomated checkChecker configured

   The document artifact parses as JSON and satisfies the Problem's document_schema.

   A conforming checker is configured. Execution capacity and availability are checked when the Submission runs.

   TfP builtin checker

   This checker executes code inside the TfP Worker. It is not a pullable OCI image.

   Read checker code

   Source provenance and runtime metadata

   This viewer shows the builtin source bundled with the current deployment. It is not a historical execution or build attestation.

   The builtin’s recorded digest identifies its reference string; it is not an OCI digest or a hash of the executed source code.

   Recorded image / builtin reference

   tfp-builtin/json-schema-check:1

   Recorded runtime digest

   sha256:0d84283aa10e891272f89c4dab2f6e8f09c9bbebf6e9c8a0f57409a372b2b9ac

   Declared entrypoint (argument array)

   ["/tfp/verify"]

   Declared source

   Open recorded source repository ↗

   No exact source commit is recorded.

   Declared license: MIT. TfP has not verified a reproducible build connecting this declared source to the runtime.

   Checker identity and execution limits

   Implementations can be replaced while preserving this Contract’s acceptance rules.

   {
     "verifier_implementation_id": "vfi_koilkata2b",
     "version": 1,
     "content_hash": "sha256:71f3458240b91bead5b735f1ebfd674d443efc6ac299485663c2bb388fab266f",
     "conformance": {
       "attempts": 1,
       "checked_at": "2026-10-13T17:54:39Z",
       "differential_against_implementation_id": null,
       "failures": [],
       "fixtures_passed": 2,
       "fixtures_total": 2,
       "last_attempted_at": "2026-10-13T17:54:39Z",
       "last_execution_error": null,
       "status": "passed"
     },
     "runtime_image_digest": "sha256:0d84283aa10e891272f89c4dab2f6e8f09c9bbebf6e9c8a0f57409a372b2b9ac",
     "limits": {
       "disk_mb": 64,
       "memory_mb": 512,
       "network": "none",
       "timeout_seconds": 60,
       "vcpu": 1
     }
   }

   Stage evidence schema

   {
     "properties": {
       "valid": {
         "type": "boolean"
       }
     },
     "required": [
       "valid"
     ],
     "type": "object"
   }

2. Campaign reviewer judgmentReviewReview required

   One campaign reviewer reads the report, inspects the code and the result document, re-runs the code where the Problem statement asks for it, and judges the Submission against the statement's acceptance bar.

   Eligible reviewers must complete the pinned rubric. Seat availability and completion are not guaranteed.

   Read the pinned rubric and reviewer rules

   OCx25 acceptance review · pinned protocol

   Campaign reviewer · 1 reviewer

   Judge the Submission against the acceptance bar in the Problem statement, nothing else.

   1. Read the report and the result document. Check that the targets, folds and metric are the Campaign's (the ocx25-data Skill: prepare_targets.py targets, leave-one-composition-out by element system, Pearson R² of pooled out-of-fold predictions from evaluate.py). A Submission that changed the evaluation fails, however good its numbers.
   2. Where the statement asks for a re-run, unpack the code archive and run it as the document's reproduction.command says. Re-score every predictions table named in the document with evaluate.py against targets you derived yourself from the bound inputs. Set reran_code to true only if you did this. Numbers that do not reproduce within rounding fail.
   3. Read the leakage audit and the feature sources. Any path by which a held-out element system's electrochemical outcomes (voltages, Faradaic efficiencies, production rates, or anything derived from them) could reach training, feature construction or model selection fails the Submission.
   4. Answer pass when the bar is met, fail when it is not, uncertain only when something outside the Submission prevents you from deciding (say what). Cite the files and lines your rationale relies on.

   Rationale requiredAt least 1 citationConfidence required

   Response fields and full schema

   • reran_code · required boolean
   • verdict · required "pass" / "fail" / "uncertain"

   {
     "additionalProperties": false,
     "properties": {
       "reran_code": {
         "type": "boolean"
       },
       "verdict": {
         "enum": [
           "pass",
           "fail",
           "uncertain"
         ]
       }
     },
     "required": [
       "verdict",
       "reran_code"
     ],
     "type": "object"
   }

   Only designated Campaign reviewers may claim a seat.The submitter is excluded.Responses on this node are withheld until it completes.Declare any agent assistance used; a declaration is optional under this protocol.

   Exact reviewer eligibility, visibility and Skill pins

   {
     "reviewers": {
       "allowed": null,
       "assistance": {
         "allowed": [],
         "required": false
       },
       "count": 1,
       "eligibility": {
         "organization_id": null,
         "rule": "campaign_reviewers",
         "user_ids": null
       },
       "independent": true,
       "kind": "human"
     },
     "conflicts": {
       "exclude_submitter": true,
       "exclude_submitter_organization": false
     },
     "visibility": {
       "hidden": [],
       "prior_nodes": [],
       "stage_evidence": true
     },
     "skills": [
       {
         "requirement": "recommended",
         "skill_version_id": "skv_pusvdzhbcm"
       }
     ]
   }

   How judgments become the outcome

   Decide

   Accept when at least 1 responses have verdict equals "pass". Reject when at least 1 have verdict equals "fail". Otherwise unresolved.

   Inputs, in order: reviewer · decision

   Final outcome comes from decision. Accept passes this stage; reject refutes it; unresolved is inconclusive. All verification stages must pass.

   Node deadline: 336 hoursClaim hold: 1440 minutes, bounded by the node deadline

   Protocol identity and full rules

   rpt_cci7slea7s
sha256:fbf20eed94012919eb3647563f929c392706ecca5903dbba5532448169b87f3c

   Read the complete pinned protocol →

   {
     "output": "decision",
     "policy": {
       "assignment_timeout_hours": 336,
       "claim_hold_minutes": 1440
     }
   }