메뉴
Audits dicom_series_preflight evidence packs for pass/warn/fail preflight verdicts with explicit corruption, orientation, PHI, and consistency findings.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Grades a skill or verifier directory for required files, valid manifest structure, side-effect declarations, validation gates, paired-verifier resolution, fixtures, and authoring hygiene. Engineering verification only.
Deterministic verifier for nv_segment_ct_finetune evidence packs. Audits the finetuned checkpoint, training trajectory, and dataset audit signals already recorded by the wrapper. Engineering verification only.
Audits nv_segment_ct evidence packs for label-map readability, anatomy plausibility (organ volume bounds, fragmentation, bilateral symmetry, liver-larger-than-spleen), and optional per-class Dice/IoU against a referenced ground-truth label map. Engineering verification only.
Deterministic paired verifier for nv_generate_ct_rflow evidence packs. Audits per-sample image/mask geometry, CT-HU plausibility, mask label-set sanity, declared output-label coverage, and lung-lobe HU plausibility. CPU-only; does not re-run synthesis.
Used to verify dicom-metadata-extract evidence packs for source gate success, header fact completeness, PHI-flag consistency, and scope disclosure. Not for de-identification.
Used to verify dicom-series-to-volume evidence packs by comparing JSON geometry evidence with the emitted NIfTI artifact. Not for clinical validation.
| name | dicom-preflight-quality-v1 |
| description | Audits dicom_series_preflight evidence packs for pass/warn/fail preflight verdicts with explicit corruption, orientation, PHI, and consistency findings. |
| license | Apache-2.0 |
dicom_series_preflight evidence packs for explicit pass/warn/fail preflight findings.dicom_series_preflight_evidence_pack; outputs are dicom_preflight_quality_report.scripts/grade.py on the preflight evidence-pack directory.run_script, use run_script("scripts/grade.py", args=["RUNS/PREFLIGHT_PACK"]).| Script | Purpose | Arguments |
|---|---|---|
scripts/grade.py | Primary verifier entrypoint declared by skill_manifest.yaml. | EVIDENCE_PACK_DIR |
output.json and validation_summary.json.| Error | Cause | Fix |
|---|---|---|
| Missing pack JSON | The fixture path is not a completed evidence pack. | Point to the source preflight run directory. |
| Unexpected fail verdict | The source preflight pack recorded fail-level findings. | Inspect preflight_gate.findings in the verifier output. |
| No findings | The source wrapper did not emit preflight details. | Re-run the source skill and preserve output.json. |
Paired verifier for skills/dicom-series-preflight.
python verifiers/dicom_preflight_quality_v1/scripts/grade.py runs/preflight_trusted
make run-trusted SKILL=dicom_series_preflight \
FIXTURE=skills/dicom-series-preflight/fixtures/clean_no_phi \
OUT=runs/preflight_trusted
Reads the target pack's output.json and validation_summary.json. Emits
overall of pass, warn, or fail plus a preflight_gate.findings list.
Not for clinical de-identification or regulatory clearance.