How do I install the Dicom Metadata Quality V1 skill?

Run npx skillmds@latest add nvidia-medtech/dicom-metadata-quality-v1 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Dicom Metadata Quality V1 safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: executes scripts. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Dicom Metadata Quality V1?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Dicom Metadata Quality V1 free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under Apache-2.

Who published Dicom Metadata Quality V1?

NVIDIA-Medtech (@nvidia-medtech) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Dicom Metadata Quality V1 by nvidia-medtech

Name: Dicom Metadata Quality V1
Rating: 4.5 (4 reviews)
Author: NVIDIA-Medtech

File contents

DICOM Metadata Quality Verifier

Purpose

Used for deterministic second-pass review of a dicom-metadata-extract evidence pack.
Checks that the source pack passed, required header facts are present, PHI tag presence is represented consistently, and the limited PHI scope is disclosed.
Reports standard PHI tag presence as a warning, not a failure.
Manifest I/O: inputs are dicom_metadata_evidence_pack; outputs are dicom_metadata_quality_report.

Instructions

Use this verifier only on an evidence pack directory produced by skills/dicom-metadata-extract.
Run it through eval_engine/run.py when producing verifier evidence.
The verifier entrypoint is scripts/grade.py; do not reimplement its checks in an agent prompt.
Do not treat a pass or warning as de-identification proof.

Available Scripts

Script	Purpose	Arguments
`scripts/grade.py`	Primary verifier entrypoint declared by `skill_manifest.yaml`.	`EVIDENCE_PACK_DIR`

Prerequisites

No network, GPU, Docker, or extra Python package dependencies are required.
The input directory must contain manifest.json, validation_summary.json, and output.json from a dicom-metadata-extract evidence pack.

Limitations

PHI tag presence is advisory; this is not a de-identification audit.
Private tags and burnt-in pixel text remain out of scope.
The verifier audits the emitted metadata facts; it does not re-read the original DICOM file.

Troubleshooting

Error	Cause	Fix
`target_skill_matches` fails	The fixture is not a `dicom-metadata-extract` evidence pack.	Re-run the verifier against the source skill pack directory.
`source_pack_passed` fails	The source skill gates did not pass.	Inspect the source pack's `validation_summary.json` before verifier findings.
`phi_scope_disclosed` fails	The source output no longer discloses standard-tag, private-tag, or burnt-in pixel limits.	Restore explicit scope text in the source skill output.

Example

python verifiers/dicom_metadata_quality_v1/scripts/grade.py \
  examples/evidence_packs/dicom_metadata_pass

python eval_engine/run.py verifiers/dicom_metadata_quality_v1 \
  --fixture examples/evidence_packs/dicom_metadata_pass \
  --out runs/dicom_metadata_quality

---name: dicom-metadata-quality-v1description: 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.license: Apache-2.0---# DICOM Metadata Quality Verifier## Purpose- Used for deterministic second-pass review of a `dicom-metadata-extract` evidence pack.- Checks that the source pack passed, required header facts are present, PHI tag presence is represented consistently, and the limited PHI scope is disclosed.- Reports standard PHI tag presence as a warning, not a failure.- Manifest I/O: inputs are `dicom_metadata_evidence_pack`; outputs are `dicom_metadata_quality_report`.## Instructions- Use this verifier only on an evidence pack directory produced by `skills/dicom-metadata-extract`.- Run it through `eval_engine/run.py` when producing verifier evidence.- The verifier entrypoint is `scripts/grade.py`; do not reimplement its checks in an agent prompt.- Do not treat a pass or warning as de-identification proof.## Available Scripts| Script | Purpose | Arguments ||---|---|---|| `scripts/grade.py` | Primary verifier entrypoint declared by `skill_manifest.yaml`. | `EVIDENCE_PACK_DIR` |## Prerequisites- No network, GPU, Docker, or extra Python package dependencies are required.- The input directory must contain `manifest.json`, `validation_summary.json`, and `output.json` from a `dicom-metadata-extract` evidence pack.## Limitations- PHI tag presence is advisory; this is not a de-identification audit.- Private tags and burnt-in pixel text remain out of scope.- The verifier audits the emitted metadata facts; it does not re-read the original DICOM file.## Troubleshooting| Error | Cause | Fix ||---|---|---|| `target_skill_matches` fails | The fixture is not a `dicom-metadata-extract` evidence pack. | Re-run the verifier against the source skill pack directory. || `source_pack_passed` fails | The source skill gates did not pass. | Inspect the source pack's `validation_summary.json` before verifier findings. || `phi_scope_disclosed` fails | The source output no longer discloses standard-tag, private-tag, or burnt-in pixel limits. | Restore explicit scope text in the source skill output. |## Example```bashpython verifiers/dicom_metadata_quality_v1/scripts/grade.py \ examples/evidence_packs/dicom_metadata_passpython eval_engine/run.py verifiers/dicom_metadata_quality_v1 \ --fixture examples/evidence_packs/dicom_metadata_pass \ --out runs/dicom_metadata_quality```

Frequently asked questions

Dicom Metadata Quality V1