{
  "category": "work",
  "slug": "work-classify-mailroom-documents",
  "title": "AI Mailroom Classification Across Five Document Types — Completed Benchmark Result: 8/10",
  "task": "classify incoming business documents for the mailroom",
  "excerpt": "The completed WFT-034 synthetic field test reached 8/10 after one failure-only correction: 4 of five static checks passed for Document Classification, while 1 check remained unresolved.",
  "tool": "Codex multi-agent session",
  "model": "Exact underlying model identifier not disclosed by the Codex session",
  "publishedAt": "2026-06-09T08:00:00+08:00",
  "durationMinutes": 0,
  "testMode": "Synthetic benchmark",
  "inputDisclosure": "Synthetic blind-test input pack WFT-034: A shared-services team will provide synthetic invoices, notices, applications, correspondence, and ambiguous document types. Source facts: six fictional records WFT-034-C01 through WFT-034-C06; policy rules P1–P5; scores 45, 53, 65, 62, 74, and 91; a shared surname on C02/C05; and an incomplete identifier on WFT-034-C04. Governing rule card: all five written policy rules without adding an unstated tie-breaker. Apply the written rules in order; use descending disclosed scores only where ranking is requested; a shared name alone never proves identity; incomplete identifiers require abstention rather than invention. The pack deliberately contains no answer key, target ranking, expected classification, expected calculation result, or row-level decision. Every person, organization, record, statement, file, and identifier is fictional; no private, production, or live data was supplied.",
  "runDisclosure": "We ran one text-only synthetic benchmark WFT-034 for “classify incoming business documents for the mailroom” in a Codex multi-agent session. We froze the first response, returned only its 2 detected check failures, accepted one corrected response, and scored both against the same five static checks. No external or live action occurred: nothing was sent, published, deployed, uploaded, submitted, purchased, booked, contacted, emailed, called, messaged, executed, or changed outside the fictional text fixtures. No external, live, or production action occurred. Per-case elapsed time was not instrumented during the batch session.",
  "prompt": "Run bounded synthetic field test WFT-034. Task: classify incoming business documents for the mailroom. Context: A shared-services team will provide synthetic invoices, notices, applications, correspondence, and ambiguous document types. Fictional source facts: six fictional records WFT-034-C01 through WFT-034-C06; policy rules P1–P5; scores 45, 53, 65, 62, 74, and 91; a shared surname on C02/C05; and an incomplete identifier on WFT-034-C04. Governing policy, formula, or rubric: all five written policy rules without adding an unstated tie-breaker. Apply the written rules in order; use descending disclosed scores only where ranking is requested; a shared name alone never proves identity; incomplete identifiers require abstention rather than invention. Produce a record-by-record decision matrix, ranked queue, and abstention log. Derive every row from the source facts and rule card; show calculations or criterion paths, cite supplied identifiers, state assumptions, surface any violated constraint, ambiguity, missing datum, or source conflict without presuming which row should pass, and abstain where evidence is incomplete. Do not infer an expected answer from the evidence plan and do not claim any live or external action. Planned verification after the frozen response: A routing index compared with held-back document labels will verify category and destination accuracy.",
  "firstResult": "Frozen first response WFT-034 produced a record-by-record decision matrix, ranked queue, and abstention log for the task “classify incoming business documents for the mailroom.” It treated the supplied pack as fictional and proposed this central handling: rank WFT-034-C06 first under P2, keep C02 and C05 separate despite the shared surname, and abstain on WFT-034-C04 until its identifier can be resolved. Concrete saved artifact row WFT-034-ROW1 reads: “WFT-034-C01 | rank WFT-034-C06 first under P2, keep C02 and C05 separate despite the shared surname, and abstain on WFT-034-C04 until its identifier can be resolved | source: supplied fictional pack | review state: first-draft.” The draft preserved the named records and separated supplied facts from assumptions; its inspectable work product also stated that no message, transaction, system change, or learner outcome had occurred. On the five predeclared static checks, it passed Document Classification rule accuracy [WFT-034], Document Classification exception handling [WFT-034], and Document Classification source traceability [WFT-034]. The audit found concrete failures: for Document Classification task fidelity [WFT-034], the saved draft did not connect WFT-034-C04 to the full boundary of “classify incoming business documents for the mailroom”; for Document Classification handoff usability [WFT-034], the saved draft left the record-by-record decision matrix, ranked queue, and abstention log without a complete reviewer handoff and acceptance marker. Those were evidence defects in the saved response, not inferred real-world failures, and they became the complete boundary of the single correction pass.",
  "correctionPrompt": "Correct only these detected WFT-034 first-draft failures, using no new input or goal: 1) Document Classification task fidelity [WFT-034] — the draft did not connect WFT-034-C04 to the full boundary of “classify incoming business documents for the mailroom”; 2) Document Classification handoff usability [WFT-034] — the draft left the record-by-record decision matrix, ranked queue, and abstention log without a complete reviewer handoff and acceptance marker.",
  "finalResult": "Corrected response WFT-034 retained the original fictional inputs, task boundary, and central decision: rank WFT-034-C06 first under P2, keep C02 and C05 separate despite the shared surname, and abstain on WFT-034-C04 until its identifier can be resolved. Concrete corrected artifact row WFT-034-ROW1 reads: “WFT-034-C01 | rank WFT-034-C06 first under P2, keep C02 and C05 separate despite the shared surname, and abstain on WFT-034-C04 until its identifier can be resolved | evidence locator: WFT-034-C01 | static status: 8/10 after one correction.” It changed only the detected failure areas, adding inspectable rows for Document Classification handoff usability [WFT-034]. The frozen final text passed Document Classification rule accuracy [WFT-034], Document Classification exception handling [WFT-034], Document Classification source traceability [WFT-034], and Document Classification handoff usability [WFT-034] and still failed Document Classification task fidelity [WFT-034]. The final record-by-record decision matrix, ranked queue, and abstention log therefore earned 8/10 from 4 passing checks; no second repair was attempted. This result reports only a bounded transcript-and-fixture evaluation. It does not claim that any workplace process, learner performance, account, device, service, or external system was actually changed, contacted, tested live, or improved.",
  "checks": [
    {
      "name": "Document Classification task fidelity [WFT-034]",
      "firstPass": false,
      "finalPass": false,
      "evidence": "WFT-034 static check 1 inspected the saved wording for “Document Classification task fidelity [WFT-034].” The first transcript did not contain enough support; the corrected transcript did not contain enough support. The comparison used only fictional fixture WFT-034-C04, the declared Document Classification rules, and the record’s frozen text; no live outcome or external action counted as evidence."
    },
    {
      "name": "Document Classification rule accuracy [WFT-034]",
      "firstPass": true,
      "finalPass": true,
      "evidence": "WFT-034 static check 2 inspected the saved wording for “Document Classification rule accuracy [WFT-034].” The first transcript contained enough support; the corrected transcript contained enough support. The comparison used only fictional fixture WFT-034-C04, the declared Document Classification rules, and the record’s frozen text; no live outcome or external action counted as evidence."
    },
    {
      "name": "Document Classification exception handling [WFT-034]",
      "firstPass": true,
      "finalPass": true,
      "evidence": "WFT-034 static check 3 inspected the saved wording for “Document Classification exception handling [WFT-034].” The first transcript contained enough support; the corrected transcript contained enough support. The comparison used only fictional fixture WFT-034-C04, the declared Document Classification rules, and the record’s frozen text; no live outcome or external action counted as evidence."
    },
    {
      "name": "Document Classification source traceability [WFT-034]",
      "firstPass": true,
      "finalPass": true,
      "evidence": "WFT-034 static check 4 inspected the saved wording for “Document Classification source traceability [WFT-034].” The first transcript contained enough support; the corrected transcript contained enough support. The comparison used only fictional fixture WFT-034-C04, the declared Document Classification rules, and the record’s frozen text; no live outcome or external action counted as evidence."
    },
    {
      "name": "Document Classification handoff usability [WFT-034]",
      "firstPass": false,
      "finalPass": true,
      "evidence": "WFT-034 static check 5 inspected the saved wording for “Document Classification handoff usability [WFT-034].” The first transcript did not contain enough support; the corrected transcript contained enough support. The comparison used only fictional fixture WFT-034-C04, the declared Document Classification rules, and the record’s frozen text; no live outcome or external action counted as evidence."
    }
  ],
  "initialScore": 6,
  "score": 8,
  "verdict": "worked",
  "recommended": true,
  "whatWorked": [
    "WFT-034 kept “classify incoming business documents for the mailroom” bounded to disclosed fictional inputs and preserved an auditable first-response snapshot.",
    "WFT-034 made the central handling—rank WFT-034-C06 first under P2, keep C02 and C05 separate despite the shared surname, and abstain on WFT-034-C04 until its identifier can be resolved—inspectable rather than implying unseen work.",
    "WFT-034 earned final passes for Document Classification rule accuracy [WFT-034] and Document Classification exception handling [WFT-034] under the same frozen scoring rules."
  ],
  "whatFailed": [
    "WFT-034 still lacked enough saved-text evidence for Document Classification task fidelity [WFT-034]; the record leaves that final failure visible."
  ],
  "evidencePlan": "A routing index compared with held-back document labels will verify category and destination accuracy.",
  "evidenceNotes": [
    "WFT-034 preserves the exact synthetic prompt, frozen first-response account, failure-only correction, corrected-response account, and five boolean decisions together.",
    "WFT-034 scores are arithmetic: 3 first-pass checks × 2 = 6/10; 4 final-pass checks × 2 = 8/10.",
    "WFT-034 evaluated only the text/static portion of the declared evidence plan—A routing index compared with held-back document labels will verify category and destination accuracy.—and did not fabricate a live artifact, external validator, or observed outcome."
  ],
  "limitations": [
    "WFT-034 is a synthetic benchmark, so its worked verdict measures fit to the disclosed fictional Document Classification fixtures rather than effectiveness in a real workplace or learning setting.",
    "WFT-034 used one text-only Codex multi-agent session whose underlying model identifier was not disclosed; different prompts, models, fixtures, or human reviewers could produce different results."
  ]
}