# Frontier Full-OfOne Mode A Controlled Execution Contract

Status: `executed_reviewed`
Protocol: `research/frontier-full-ofone-repair-protocol.md`
Mode: `Mode A: Controlled Non-Deep-Research Execution`
Prepared: `2026-05-21T03:02:51-06:00`
Executed: `2026-05-21T03:34:00-06:00`

This contract was the authorized local/operator-driven repair path for the unrepaired frontier full-OfOne strategic gated diligence repeat-1 slot. It is not a launched ChatGPT Deep Research run. Its controlled Mode A rerun 5 output is now completed, validator-valid, locally reviewed, and matrix-inserted only as replacement evidence for aggregate scoring.

## Run Metadata

Batch ID: `2026-05-17-batch-01`
Case ID: `case-strategic-gated-diligence-001`
Arm ID: `full_ofone`
Model family: `frontier_reasoning`
Repeat: `1`
Run ID: `2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5`
Rerun of: `2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1`
Rerun number: `5`
Status before execution: `prepared_not_executed`
Aggregate policy before execution: `not_aggregate_eligible`
Status after execution: `reviewed`
Aggregate policy after execution: `replace_for_aggregate_only`

## Source Binding

Use only the frozen benchmark inputs listed here:

- Case file: `benchmarks/cases/strategic-gated-diligence.md`
- Case file SHA-256: `sha256:18a0247003e142c80c8748eb4652f900b81ca409b91ba7c370e1362d24680942`
- Arm prompt: `benchmarks/runs/2026-05-17-batch-01/prompts/full_ofone.md`
- Arm prompt SHA-256: `sha256:613afac8909b34accb57fd2c24217bb59a28a45860e32f36f6ec5f7f4ab5587e`
- Input bundle SHA-256: `sha256:4168a4e6533f1398611d704254b48c8fbfde5f547a4a0c79cd072a98fdbacd44`

Canonical artifact-level `benchmark_trace` values:

```json
{
  "case_id": "case-strategic-gated-diligence-001",
  "run_id": "2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5",
  "case_file": "benchmarks/cases/strategic-gated-diligence.md",
  "case_file_sha256": "sha256:18a0247003e142c80c8748eb4652f900b81ca409b91ba7c370e1362d24680942",
  "prompt_file": "benchmarks/runs/2026-05-17-batch-01/prompts/full_ofone.md",
  "prompt_file_sha256": "sha256:613afac8909b34accb57fd2c24217bb59a28a45860e32f36f6ec5f7f4ab5587e",
  "input_bundle_sha256": "sha256:4168a4e6533f1398611d704254b48c8fbfde5f547a4a0c79cd072a98fdbacd44"
}
```

## Required Raw Output Shape

The final raw Markdown output must begin with this exact first heading:

```markdown
# Benchmark Raw Output
```

It must then include exact run metadata with:

```markdown
Run ID: `2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5`
Rerun of: `2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1`
Status: `completed`
```

It must include actual case-bound sections:

- `## Artifact JSON`
- `## Validator Result`
- `## Rendering`
- `## Patch Report`

Do not return a memo, report, analysis of this contract, research narrative, or advisory plan. Return only the benchmark package.

## Controlled Execution Rules

- Do not use attachments as the execution contract.
- Do not launch another same-shape ChatGPT Deep Research remedial rerun.
- Do not inspect outputs from other benchmark arms while authoring the replacement.
- Do not reuse the excluded original or any failed remedial artifact as the replacement artifact.
- Do not insert this run into `remedial_runs` before the required output package exists and passes computed local validation.
- Do not grant aggregate eligibility, compare aggregates, or make superiority claims before local review and Pages parity pass.

## Required Local Verification

After a controlled output package exists, run:

```bash
npm run frontier:controlled:check
npm run frontier:protocol:check
npm run schema:check
npm run validate
npm run review:check
npm run research:check
npm run benchmark
npm test
```

After commit and push, run:

```bash
npm run pages:check
```

Superiority claims remain blocked.

## Execution Result

Controlled rerun 5 was produced under this contract and preserved as separate raw, machine, and review artifacts:

- Raw output: `benchmarks/runs/2026-05-17-batch-01/outputs/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.md`
- Artifact JSON: `benchmarks/runs/2026-05-17-batch-01/outputs/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.artifact.json`
- Validator JSON: `benchmarks/runs/2026-05-17-batch-01/outputs/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.validator.json`
- Rendering: `benchmarks/runs/2026-05-17-batch-01/outputs/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.rendering.md`
- Patch report: `benchmarks/runs/2026-05-17-batch-01/outputs/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.patch.json`
- Local review: `benchmarks/reviews/2026-05-17-batch-01/2026-05-17-batch-01__case-strategic-gated-diligence-001__full_ofone__frontier_reasoning__r1__rerun5.md`

Pinned SHA-256 values:

- Raw output: `sha256:2897a11f82baee33a0161a8d8f64b716017c3fedd9d22a2639c60caa6ad4bf86`
- Artifact JSON: `sha256:6bd70e20ba36bf9e63a8bc4c65127a3f1166d30590d1a63df63b2612261a14f0`
- Validator JSON: `sha256:de1bbeefa04a96f0da9fb2210290c206e8c421139af5abaa43c2aa6b8850093a`
- Rendering: `sha256:116da40a3d99f20b7bce459f9de543e4e4ab0edd87c7ea5c95ee364899d211e5`
- Patch report: `sha256:60c289babea552802569575f5a4ff40672835996739841753ff20d3cdcd1ce19`
- Local review: `sha256:67c79c6fc30535dcb143ae47620b79444186d6ccc7b44087f4c65fc8789f7382`

The original frontier full-OfOne run remains excluded and immutable. Rerun 5 is not a new predeclared slot; it is replacement evidence for aggregate scoring only after validation, local review, publication, and Pages parity.