I4h Workflow Dataset Mimic

NVIDIAOfficialCreated by Isaac for Healthcare TeamDistributed by NVIDIA

Skill
Workflow
Data and AI
Apache-2.0
Medium risk

From the author

Expand workflow HDF5 demonstrations with action jitter, optionally scoped to node segments. Use for synthetic variants; do not use to collect data, alter state directly, or generate new images.

— the description by Isaac for Healthcare Team

AutoGineer summary

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

This skill provides a structured workflow for augmenting existing HDF5 demonstration datasets used in healthcare robotics workflows. It enables users to create synthetic variants of successful episodes by applying controlled action jitter, which can be scoped to specific node segments. The process involves inspecting source data, generating new episodes with configurable noise levels, and verifying the integrity of the resulting dataset.

What it does

Inspects HDF5 demonstration files to identify successful episodes and segments.
Generates synthetic dataset variants by applying action jitter to source demonstrations.
Supports node-scoped jitter application for targeted data augmentation.
Verifies generated datasets for consistency in dimensions and success flags.
Integrates with visualization tools for post-generation review.

When to use it

When you need to increase the volume of demonstration data for training models.
When you want to introduce controlled variability into existing successful robot episodes.
When you need to test model robustness by adding noise to specific action nodes.
When preparing datasets for further visualization or analysis workflows.

Advantages

Automates the creation of synthetic training data from existing successful runs.
Allows for precise control over noise levels and random seeds.
Supports targeted augmentation via node-specific segment scoping.
Includes built-in verification steps to ensure data integrity.

Usage examples

“Mimic 3 more episodes and visualize my dataset.”: The agent generates three synthetic variants of the source HDF5 file with action jitter, verifies the output, converts the data, and launches the LeRobot visualizer.
“Create 10 new episodes from my source file with 0.05 noise.”: The agent creates 10 additional episodes in a new directory using the specified noise level and reports the final dataset counts.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/nvidia/i4h-workflow-dataset-mimic/download

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

mkdir -p .agents/skills && unzip i4h-workflow-dataset-mimic.zip -d .agents/skills

Documentation by Isaac for Healthcare Team

Mimic Workflow Demonstrations

Purpose

Clone successful source episodes and perturb recorded actions.

Instructions

Run the checkout resolver and select the source HDF5.
Inspect successful episodes and segments.
Run mimic to generate the requested additional action-jitter variants.
Compare exact counts and preserved contracts.

Resolve and inspect

export I4H_WORKFLOWS_REPO_URL="${I4H_WORKFLOWS_REPO_URL:-https://github.com/isaac-for-healthcare/i4h-workflows}"
I4H_REPO_DIR_NAME="${I4H_WORKFLOWS_REPO_URL%/}"
I4H_REPO_DIR_NAME="${I4H_REPO_DIR_NAME##*/}"
I4H_REPO_DIR_NAME="${I4H_REPO_DIR_NAME##*:}"
I4H_REPO_DIR_NAME="${I4H_REPO_DIR_NAME%.git}"
[ -n "$I4H_REPO_DIR_NAME" ] || { echo "Cannot derive a checkout name from I4H_WORKFLOWS_REPO_URL" >&2; exit 2; }
ROOT="${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>/dev/null)}"
if [ ! -d "$ROOT/workflows/i4h_workflows" ]; then
  ROOT="${I4H_WORKFLOWS:-$HOME/$I4H_REPO_DIR_NAME}"
  [ -d "$ROOT/workflows/i4h_workflows" ] || git clone "$I4H_WORKFLOWS_REPO_URL" "$ROOT"
fi
export I4H_WORKFLOWS="$ROOT"
cd "$ROOT"
HDF5_PATH=/absolute/path/to/source.hdf5
uv run --project tools/dataset i4h-dataset inspect "$HDF5_PATH" --segments

Treat the resolver above as part of the skill contract: a hosted copy may run outside the base repository, so never assume the current checkout contains workflows/i4h_workflows. I4H_WORKFLOWS_REPO_URL selects the clone source. When I4H_WORKFLOWS is unset, derive the fallback directory from that URL; set I4H_WORKFLOWS only to reuse or choose a specific destination. Never replace an existing checkout.

Use the explicit/current-chain recording. Require at least one successful source episode. If the user asks for node-scoped jitter, require that node in segment metadata.

Expand

RUN_DIR="$(pwd)/runs/<workflow>/$(date +%Y%m%d_%H%M%S)"
mkdir -p "$RUN_DIR"
uv run --project tools/mimic i4h-mimic \
  "$HDF5_PATH" "$RUN_DIR/expanded.hdf5" \
  --episodes <additional-count> \
  --noise 0.01 \
  --seed <seed> \
  --include-source \
  --successful-only

--episodes is the number of generated variants, not the final total. Add --node <node-id> only when requested. Preserve the default deterministic seed unless the user supplies another.

Verify and compose

uv run --project tools/dataset i4h-dataset inspect "$RUN_DIR/expanded.hdf5" --segments

Confirm source count, generated count, dimensions, success flags, and segment preservation. Reject zero generated episodes.

If the same prompt requests visualization, continue with i4h-workflow-dataset-convert, then i4h-lerobot-viz; raw HDF5 is not a LeRobot visualizer input.

Troubleshooting

On failure, inspect the first episode shape or segment error. If a requested node is absent, report available node ids.

Prerequisites

Require a readable HDF5 file with at least one successful episode and valid segments for node-scoped jitter.

Limitations

The current tool does not independently add state noise, rerun physics, or synthesize camera frames.

Examples

Mimic 3 more episodes and visualize my dataset. → generate three variants, convert the expanded HDF5, then launch LeRobot visualization.

Completion gate

Report input/output paths, source and generated counts, final total, action-noise value, seed, optional node, limitations, and conversion/visualizer result when requested.

Files

ATTRIBUTION.mdAdded by AutoGineer955 B
BENCHMARK.md6.8 KiB
LICENSELicence file added by AutoGineer11.2 KiB
SKILL.md3.8 KiB
evals/evals.json1.3 KiB
skill-card.md4.4 KiB
skill.oms.sig4.5 KiB

BENCHMARK.md

Skill Benchmark: i4h-workflow-dataset-mimic

✅ Overall verdict: PASS — Recommended for publication

Publication Recommendation

Recommended for publication based on the completed evaluation evidence in this report.

Evaluation Metadata

Skill: i4h-workflow-dataset-mimic
Evaluation date: 2026-09-15
Evaluator version: 1.5.6
Agents: Claude Code (aws/anthropic/bedrock-claude-opus-4-8), Codex (openai/openai/gpt-5.5)
Tasks: 2 evaluation tasks (2 positive)
Dataset digest: sha256:26e64d6309ea0d85ebda82fbe5e6c44c2789e399a1a2bbf4ffc90b73dd488c46 (skill-evaluator-dataset-snapshot/1)
Attempts per task: 3
Environment: k8s-sandbox
Tier 2 evidence: required for publication
Tier 3 evidence: required for publication

Each task attempt ran in its own isolated sandbox pod.

What This Report Answers

The three-tier evaluation checks whether the skill:

is safe to use;
produces correct answers;
is discovered and activated when needed;
helps the agent complete the user's goal and expected workflow; and
avoids wasted skill and tool usage.

Results at a Glance

Measure	Claude Code (Baseline → Skill Uplift)	Codex (Baseline → Skill Uplift)
Overall	88.1% — baseline ran, but no comparable score was available; uplift unavailable	70.3% — baseline ran, but no comparable score was available; uplift unavailable
Security	100.0% → 100.0% (±0.0 points)	83.3% → 50.0% (-33.3 points)
Correctness	8.0% → 100.0% (+92.0 points)	13.3% → 90.0% (+76.7 points)
Discoverability	97.5% — baseline ran, but no comparable score was available; uplift unavailable	89.0% — baseline ran, but no comparable score was available; uplift unavailable
Effectiveness	11.0% → 57.5% (+46.5 points)	5.8% → 35.0% (+29.2 points)
Efficiency	85.5% — baseline ran, but no comparable score was available; uplift unavailable	87.7% — baseline ran, but no comparable score was available; uplift unavailable

How to read this table: baseline is the same task attempted without the target skill. Scores are rounded to one decimal; threshold-adjacent values use additional precision so their displayed band matches the verdict. Uplift is derived from those displayed scores and shown in percentage points.

Example: 47.0% → 92.0% (+45.0 points) means the skill-assisted run scored 92.0%, 45.0 percentage points above its 47.0% no-skill baseline.

Token Usage

Actual Tier 3 execution usage is reported for every observed agent/case pair and both conditions.

Agent	Dataset case	With skill	Without skill	Delta	Change	Coverage
claude-code	All cases	610,244	1,367,887	N/A	N/A	skill 2/2; base 5/5
claude-code	i4h-workflow-dataset-mimic-readme-quick-three-visualize	223,176	494,115	N/A	N/A	skill 1/1; base 3/3
claude-code	i4h-workflow-dataset-mimic-table-ten-episodes	387,068	873,772	N/A	N/A	skill 1/1; base 2/2
codex	All cases	474,679	747,543	N/A	N/A	skill 2/2; base 6/6
codex	i4h-workflow-dataset-mimic-readme-quick-three-visualize	74,123	254,167	N/A	N/A	skill 1/1; base 3/3
codex	i4h-workflow-dataset-mimic-table-ten-episodes	400,556	493,376	N/A	N/A	skill 1/1; base 3/3
ALL AGENTS	Dataset aggregate	1,084,923	2,115,430	N/A	N/A	skill 4/4; base 11/11

Prompt tokens include cached reads, so total tokens are prompt + completion (cached is not added twice). The Efficiency score uses (prompt - cached) + completion. N/A means the relevant trajectory counters were not available; coverage is never estimated.

Tier Status

Tier	Purpose	Status	Evidence
Tier 1	Static validation	PASSED WITH OBSERVATIONS	11 validator(s); 2 finding(s)
Tier 2	Semantic deduplication	PASSED	2 validator(s); 0 finding(s)
Tier 3	Live agent evaluation	PASS	2 agent(s); 2 task(s)

Findings and Observations

<details> <summary>Show detailed findings and successful checks</summary>

MEDIUM SECURITY/Unknown (SQP-2): The skill's resolver script automatically clones a remote GitHub repository into the user's home directory (defaulting t (SKILL.md:32)
LOW QUALITY/quality_reliability: Inputs are used but no dedicated Inputs section is documented (team-skills/holoscan/i4h-workflows/i4h-workflow-dataset-mimic/SKILL.md)

</details>

Scoring Methodology

<details> <summary>Show dimension definitions, source signals, and thresholds</summary>

Dimension	Question	Scored signals
Security	Is it safe to use?	`security` (100%)
Correctness	Is the answer correct?	`accuracy` (100%)
Discoverability	Was the right skill loaded when needed?	`skill_execution` (100%)
Effectiveness	Did the skill help complete the task?	`goal_accuracy` (50%) + `behavior_check` (50%)
Efficiency	Did it avoid wasted tool calls and token usage?	`skill_efficiency` (50%) + `token_efficiency` (50%)

Dimension bands: PASS at 50% or above; NEUTRAL from 40% to below 50%; FAIL below 40%.
Overall Tier 3 lift: PASS at +5 points or more; FAIL at -10 points or less; values between those bands are NEUTRAL.
Overall verdict: PASS only when every configured dimension passes for at least one supported agent. Lift is reported as diagnostic evidence and does not override this gate.
The 50% attempt pass threshold is a separate per-task gate; it is not the dimension pass threshold.
Effectiveness is the equal-weight mean of goal completion (goal_accuracy) and expected workflow adherence (behavior_check).
Efficiency is 50% tool-call productivity (the backward-compatible skill_efficiency wire id) and 50% token_efficiency. Positive-case skill routing is scored under Discoverability, not Efficiency; a negative case without a routing target is N/A. N/A sources are omitted, remaining weights are renormalized, and the dimension is marked partial.

Signals present in this run:

security (Security): unsafe operations, secret leakage, and unauthorized access.
skill_execution (Skill Execution): whether the expected skill was selected, decoys were avoided, and the workflow executed.
skill_efficiency (Tool Productivity): tool-call productivity (legacy wire id; routing is scored under Discoverability).
accuracy (Accuracy): final-answer correctness against the reference answer.
goal_accuracy (Goal Accuracy): whether the user's goal was achieved.
behavior_check (Behavior Check): whether the expected workflow behavior was followed.
token_efficiency (Token Efficiency): actual uncached prompt plus completion usage (50% of Efficiency).

</details>

Freshness

Regenerate this benchmark when the skill, evaluation dataset, target agent/model, evaluator version, environment, or scoring policy changes.

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: Needs Python, Needs Node.js

Security

What our scanner noticed· Medium risk

Large base64-encoded blobs

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

Attribution and licence

Created by

Isaac for Healthcare Team

Publisher

NVIDIA (Official)

Distributed by

NVIDIA

Source

NVIDIA/skills · skills/i4h-workflow-dataset-mimic · commit fce3cca

Licence

Apache License 2.0 (redistribution allowed)

LICENSE (added by AutoGineer from LICENSE-APACHE)
ATTRIBUTION.md (written by AutoGineer)

Synced

2026-11-08

AutoGineer is not affiliated with or endorsed by Isaac for Healthcare Team or NVIDIA.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download i4h-workflow-dataset-mimic.zip