{
  "schema": "ab5d-essay-generation/1",
  "model": "Claude (submitted claude-opus-5), reviewed by OpenAI Codex",
  "displayName": "Claude (submitted model: claude-opus-5), reviewed and revised by OpenAI Codex",
  "scope": "One submitted Claude essay with independent source, script, visual, chronology and rights review",
  "essayCount": 1,
  "usage": null,
  "perEssayUsage": null,
  "usageDetail": "Claude supplied no token telemetry. Review usage is not separately measured. Submitted model identifier is retained without independent runtime verification.",
  "contributors": {
    "operator": "Phil (AB5D coordinator), via the Claude desktop application",
    "agent": "Claude in Cowork mode, Claude Agent SDK session",
    "model": "claude-opus-5",
    "modelVersion": "claude-opus-5 (session-reported exact model id; the routing label on this batch was 'claude', which is not a model identifier)",
    "tools": [
      "Art Blocks public GraphQL index over HTTP POST",
      "browser pane on the operator's computer for page rendering, text extraction, image inspection, live renderer execution, keyboard and mouse input, and sketch state sampling",
      "independent reimplementation of the artist's pseudo-random generator, run against every token hash in the edition",
      "SHA-256 over the retrieved script bytes computed in the browser using the Web Crypto API",
      "local shell and file tools for package assembly and validation"
    ],
    "humanEditorialInvolvement": "None. No human reviewed, edited or fact-checked the prose before submission. Status is draft-for-human-review.",
    "originalProseLicense": "CC0-1.0",
    "coordinatorReview": {
      "reviewer": "OpenAI Codex",
      "scope": "Full script reading, fresh project/edition/holding query, artist chapter, four original-renderer observations, #200 control checks, standard rights sources, corrected chronology and revised prose. Submitted reimplementation and broader licence census were not independently repeated.",
      "usage": null,
      "usageNote": "No per-essay metered usage available; submitted model identity retained as reported."
    }
  }
}
