Dicom Series Preflight

From the author

Used for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.

— the description by NVIDIA MedTech Team

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

The DICOM Series Preflight skill provides a mechanism to inspect DICOM series directories by analyzing headers without decoding pixel data. It generates a structured JSON output containing an inventory of the series, orientation axcodes, potential PHI flags, and a verdict of pass, warn, or fail. This tool is intended for pre-processing validation to ensure data quality before conversion or inference tasks, and it is explicitly not designed for clinical clearance, de-identification, or production ingestion.

What it does

Scans a single DICOM series directory for header-based validation.
Generates a JSON report including inventory, orientation, and findings.
Assigns a preflight verdict of pass, warn, or fail.
Identifies potential PHI flags within the headers.
Provides integration with trusted verifier workflows.

When to use it

Before running inference or conversion on DICOM data.
When you need to verify series orientation and header integrity.
To perform rapid quality checks on medical imaging datasets.
As a gatekeeper in automated medical imaging pipelines.

Advantages

Fast execution due to header-only analysis without pixel decoding.
Standardized output format for downstream automated processing.
Helps identify potential issues early in the data pipeline.
Integrates with existing workflow automation tools.

Usage examples

“Run a preflight check on the DICOM series located in /data/dicom/series_01.”: The agent executes the preflight script on the specified directory and returns a JSON report containing the series inventory, orientation axcodes, and a pass/warn/fail verdict.
“Validate the DICOM series in /data/scan_01 using the trusted preflight verifier.”: The agent runs the preflight script and pipes the output through the trusted verifier to confirm the data meets the required engineering invariants.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/nvidia/dicom-series-preflight/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip dicom-series-preflight.zip -d .agents/skills

Documentation by NVIDIA MedTech Team

Purpose

Used for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.
Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.
Manifest I/O: inputs are dicom_dir; outputs are preflight_json.

Instructions

Read skill_manifest.yaml before changing arguments, side effects, or validation gates.
Run scripts/preflight_series.py through the documented command below; keep outputs under a caller-provided run directory.
If a host agent exposes run_script, use run_script("scripts/preflight_series.py", args=[...]); otherwise run the Bash/Python command shown below.
Check the emitted JSON and paired verifier guidance before treating the run as evidence.

Available Scripts

Script	Purpose	Arguments
`scripts/preflight_series.py`	Primary entrypoint declared by skill_manifest.yaml.	`PATH_TO_DICOM_DIR`

Prerequisites

Runtime requirements: Python packages listed in runtime.side_effects.pip_packages.
NiBabel 5.4 or newer is required so extreme-oblique axes remain labeled consistently across reorientation.
Run commands from the repository root unless an existing section below says otherwise.

Limitations

Header-only; does not decode pixel data or detect burnt-in PHI.
Canonical orientation gate assumes LPS-derived CT axcodes L,P,S.
Compressed transfer syntax and multi-frame instances are warned, not decoded.
Single-directory scan; does not reconcile multiple studies in one tree.
Not for clinical deployment, regulatory de-identification, autonomous diagnosis, production ingestion without a vetted converter.

Troubleshooting

Error	Cause	Fix
Missing dependency or import error	Runtime package drift from `skill_manifest.yaml`.	Install the packages declared in the manifest or use the documented setup command.
Empty or schema-invalid output	Wrong input path, unsupported modality, or upstream failure.	Re-run with a known fixture and inspect the wrapper JSON plus stderr.
Validation gate failure	Output violated a declared engineering invariant.	Keep the failed evidence pack and use the gate message to repair inputs or wrapper code.

Scans a DICOM directory (one series per folder) without decoding pixels. Emits JSON with inventory, orientation axcodes, PHI flags, findings, and a preflight.verdict of pass, warn, or fail.

python scripts/preflight_series.py PATH_TO_DICOM_DIR

Pair with verifiers/dicom_preflight_quality_v1 for a trusted preflight pack:

make run-trusted SKILL=dicom_series_preflight \
  FIXTURE=skills/dicom-series-preflight/fixtures/clean_no_phi \
  OUT=runs/dicom_preflight_demo

Flagship workflow:

make run-workflow \
  WORKFLOW=examples/workflows/dicom_preflight_gate.yaml \
  WORKFLOW_INPUT=skills/dicom-series-preflight/fixtures/clean_no_phi \
  WORKFLOW_OUT=runs/dicom_preflight_gate

Not for de-identification, private-tag review, or clinical clearance.

Files

ATTRIBUTION.mdAdded by AutoGineer937 B
BENCHMARK.md7.4 KiB
LICENSELicence file added by AutoGineer11.2 KiB
SKILL.md3.4 KiB
evals/evals.json2.5 KiB
fixtures/generate_fixtures.pyScript2.0 KiB
scripts/preflight_series.pyScript12.6 KiB
skill-card.md4.1 KiB
skill.oms.sig5.7 KiB
skill_manifest.yaml3.6 KiB
tests/test_preflight_series.pyScript2.0 KiB
validators/output_schema.json1.5 KiB

skill-card.md

Description:

Used for header-only preflight of one DICOM series folder before conversion or inference. Not for de-identification or clinical clearance.

This skill is for research and development only.

Owner

NVIDIA

License/Terms of Use:

Apache 2.0

Use Case:

Developers and engineers use this skill to run a header-only preflight scan of a DICOM series directory, checking orientation, spacing consistency, and PHI-tag presence before conversion or model inference.

Deployment Geography for Use:

Global

Requirements / Dependencies:

Requires API Key or External Credential: [No] Credential Type(s): [None]

Do not include secrets in prompts/logs/output; use least-privilege credentials; rotate keys as appropriate.

Known Risks and Mitigations:

Risk: Review before execution as proposals could introduce incorrect or misleading guidance into skills. Mitigation: Review and scan skill before deployment.

Reference(s):

Skill Manifest
Skill Benchmark

Skill Output:

Output Type(s): [Analysis, Files] Output Format: [JSON] Output Parameters: [1D] Other Properties Related to Output: [Emits structured preflight JSON with inventory, orientation axcodes, PHI flags, findings, and a verdict of pass, warn, or fail]

Evaluation Agents Used:

Claude Code (aws/anthropic/bedrock-claude-opus-4-8)
Codex (openai/openai/gpt-5.5)

Evaluation Tasks:

3 evaluation tasks across DICOM preflight scenarios (3 positive cases), each with 3 attempts per agent in isolated sandbox pods.

Evaluation Metrics Used:

Reported benchmark dimensions:

Security: Whether the skill is safe to use: checks for unsafe operations, secret leakage, and unauthorized access.
Correctness: Whether the final answer is correct against the reference answer.
Discoverability: Whether the right skill was loaded and activated when needed.
Effectiveness: Whether the skill helped complete the user's goal, combining goal completion (50%) and expected workflow adherence (50%).
Efficiency: Whether the skill avoided wasted tool calls and token usage, combining tool-call productivity (50%) and token efficiency (50%).

Underlying evaluation signals used in this run:

security: Unsafe operations, secret leakage, and unauthorized access.
skill_execution: Whether the expected skill was selected, decoys were avoided, and the workflow executed.
accuracy: Final-answer correctness against the reference answer.
goal_accuracy: Whether the user's goal was achieved.
behavior_check: Whether the expected workflow behavior was followed.
skill_efficiency: Tool-call productivity (legacy wire id; routing is scored under Discoverability).
token_efficiency: Actual uncached prompt plus completion usage.

Evaluation Results:

Measure	Claude Code (Baseline → Skill Uplift)	Codex (Baseline → Skill Uplift)
Overall	75.3%	71.6%
Security	100.0% → 66.7% (-33.3 pts)	100.0% → 66.7% (-33.3 pts)
Correctness	2.9% → 80.0% (+77.1 pts)	2.5% → 80.0% (+77.5 pts)
Discoverability	88.3%	81.7%
Effectiveness	13.6% → 56.1% (+42.5 pts)	17.5% → 49.7% (+32.2 pts)
Efficiency	85.2%	79.8%

Skill Version(s):

0.1.0 (source: skill_manifest.yaml)

Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal team to ensure this skill meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

(For Release on NVIDIA Platforms Only) Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns here.

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Needs: Declares allowed tools, Needs Python

Security

What our scanner noticed· Medium risk

Large base64-encoded blobs
Scripts that your agent may run

This skill includes scripts that your agent may run. Read them before you use it.

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Treatment Plans
K-Dense· Community
A documentation-only tool for structuring and validating clinician-authored treatment plans.
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository56.4k
Pydicom
K-Dense· Community
A Python framework for reading, writing, and processing local DICOM medical imaging datasets and pixel data.
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository56.4k
Clinical Decision Support
K-Dense· Community
A research-only toolkit for documenting and validating clinical decision-support evaluation, governance, and evidence-profile artifacts.
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository56.4k
Pyhealth
K-Dense· Community
Builds and validates clinical machine-learning pipelines for EHR, signals, and medical imaging using PyHealth 2.0.2.
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository56.4k
Clinicaltrials Skill
OpenAIOfficial
Submit compact ClinicalTrials.gov API v2 requests for study search, metadata, enums, search areas, and field statistics. Use when a user wants concise ClinicalTrials.gov summaries
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
Link-onlyLow riskHas scriptsGitHub stars of the source repository7.8k
Nv Segment Ct
NVIDIAOfficial
A wrapper for running NVIDIA VISTA3D segmentation on CT NIfTI volumes.
Science and researchClaude CodeClaude appsOpenAI Codex+14 more
Apache-2.0Medium riskHas scriptsGitHub stars of the source repository3.9k

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download dicom-series-preflight.zip