Nvidia Ontology Management

NVIDIAOfficialCreated by NVIDIA

Skill
Workflow
Data and AI
Apache-2.0
Medium risk

From the author

Model and publish semantic definitions in Auto Ontology. Use for terms, relationships, measures, imports, and governed results—not deployment or querying.

— the description by NVIDIA

AutoGineer summary

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

This skill provides a structured workflow for modeling, inspecting, and publishing semantic definitions in Auto Ontology. It emphasizes maintaining a rigorous semantic layer by separating routine metadata edits from complex modeling tasks and governed publication. The skill guides users through discovering current meanings, defining concepts and relationships, validating SQL attributes, and performing controlled model imports or exports. It is designed for environments where semantic accuracy and durable, governed changes are critical, ensuring that all modifications are backed by evidence and verified through exact readback procedures.

What it does

Inspects and edits semantic metadata like terms, relationships, and measures.
Validates SQL attributes and semantic definitions before application.
Manages bulk model imports and exports for catalog and semantic layer updates.
Enforces publication gates to ensure changes are durable and verified.
Provides workflows for modeling concepts, relationships, and reusable measures.
Facilitates verification of persisted changes through exact readback and comparison.

When to use it

When you need to define or update business concepts, relationships, or measures in Auto Ontology.
When performing bulk updates to the semantic model via import/export.
When you need to ensure a semantic change is governed, validated, and durable.
When troubleshooting semantic model inconsistencies or validation failures.

Advantages

Ensures semantic consistency through rigorous modeling and validation workflows.
Provides a clear path for governed publication and rollback of semantic models.
Reduces risk by requiring verification of persisted changes rather than relying on API success.
Supports complex semantic structures like derived SQL attributes and relationship paths.

Usage examples

“Define a new run-grain measure for revenue in Auto Ontology.”: The agent will guide you through the modeling.md workflow, requiring you to define the expression, grain, units, and additive dimensions before applying the change.
“Promote a revised semantic model to production.”: The agent will initiate the publication.md workflow, including backing up the current state, applying the change in isolation, and performing an exact readback comparison.

Install

Download the skill as a ZIP:
https://autogineer.com/agent-skills/nvidia/nvidia-ontology-management/download

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

mkdir -p .agents/skills && unzip nvidia-ontology-management.zip -d .agents/skills

Documentation by NVIDIA

Auto Ontology management and modeling

Purpose

Inspect and change Auto Ontology's semantic layer without confusing a successful API call with correct business meaning or durable publication. The machine-readable inputs, outputs, statuses, gates, and handoffs are in runtime-contract.yaml.

Choose the focused workflow before acting:

routine inspection or metadata edits: continue below and use write-api.md;
concepts, relationships, measures, units, grain, or policies: use modeling.md;
model promotion or reusable analytical results: use publication.md.

MCP tools only read. Create, patch, import, and compile-reset go through the Next.js /api/... gateway. Governed result rows require an approved source writer; Auto Ontology has no generic result-row writeback endpoint. Request bodies and field types live in docs/openapi/auto-ontology-api.json; this skill names operations and permissions only.

Prerequisites

Use a running, authenticated Auto Ontology deployment. Use nvidia-ontology-setup when the deployment or semantic layer is not ready.

Limitations

MCP is read-only, Auto Ontology has no generic result-row writeback endpoint, and current model import and certification APIs do not prove atomic promotion. Follow the publication gates below and fail closed when exact readback is unavailable.

Auth

Scripts use an API token (x-api-key or Authorization: Bearer). Mint it in the UI (user menu → API Tokens). A token acts as its owner — a viewer's token cannot do admin things. Creating and revoking tokens requires a signed-in session; a token cannot mint another token.

Writes below need catalog:edit unless noted. 403 means the owner's role lacks that permission, not that the path is wrong.

Instructions

Discover current meaning via MCP (search_terms, get_term, get_term_columns, get_term_sql_attributes, describe_table) or REST if MCP is absent (GET /api/terms, GET /api/terms/{term_id}, GET /api/exploration/tables/{table_id}/details).
Resolve the semantic contract when meaning changes. Record source binding, identity, population, relationship cardinality, measure expression, unit, grain, denominator, time, validity, and counterexamples; see modeling.md.
State the proposed change to the user (term rename, new SQL attribute, import, and so on). Do not silently rewrite the glossary.
Validate SQL before create/update: POST /api/sql-attributes/validate. A parse failure is HTTP 422, not valid: false.
Apply the smallest write that matches the request (see write-api.md).
Compilation is not a hidden side effect. Check GET /api/semantic-compilation/status. Do not call POST /api/semantic-compilation/reset as cleanup — it is an asynchronous, destructive rebuild of every database's compiled layer, returns 202, and requires semanticCompilation:manage.
Verify the persisted change, not merely request success. Model imports require a scoped backup and exact re-export comparison; see publication.md. Then use MCP check_answerable / ask_question (or REST POST /api/question-entity-coverage and POST /api/chat/completions) for positive and negative behavior checks.

Semantic relationships and "what does this dataset mean"

Stay on the semantic layer. Do not browse raw schemas to answer meaning.

Term → columns: MCP get_term_columns or GET /api/terms/{term_id}/column-attributes
Term → derived SQL: MCP get_term_sql_attributes
Table → terms and SQL attributes: MCP describe_table or GET /api/exploration/tables/{table_id}/details
Semantic hop chain between two terms: GET /api/exploration/terms/{term_id}/path/{other_term_id}
Semantic graphs: GET /api/exploration/graph, GET /api/exploration/semantic-graph

These are semantic relationship paths, not generation provenance, source revision, certification history, or version lineage.

Bulk import / export

POST /api/model/export — YAML of catalog + semantic layer. Permission modelInterchange:export. Body may set catalog database IDs in databases (empty = all) and format (auto_ontology or ossie).
POST /api/model/import — multipart YAML; native Auto Ontology (data_layer / semantic_layer) or Apache Ossie (version, name, datasets at the root). Query replace (default true) and embed (default true). Permission modelInterchange:import. Can replace existing data — confirm with the user before replace=true, apply first in isolation, and use the exact readback workflow in publication.md.

Examples

To define a run-grain measure, follow the evidence and counterexample workflow in modeling.md.
To revise and promote a model, back up the exact scope and follow the staged readback workflow in publication.md.

Troubleshooting

For write failures, verify the authenticated owner's permission, distinguish HTTP 422 parse failures from validation results, and inspect compilation status before changing the model or resetting compilation.

Files

ATTRIBUTION.mdAdded by AutoGineer936 B
BENCHMARK.md9.7 KiB
LICENSELicence file added by AutoGineer11.2 KiB
SKILL.md6.2 KiB
assets/runtime-contract.yaml2.4 KiB
evals/evals.json10.4 KiB
references/modeling.md3.8 KiB
references/publication.md4.7 KiB
references/write-api.md5.4 KiB
skill-card.md4.6 KiB
skill.oms.sig5.4 KiB

BENCHMARK.md

Skill Benchmark: nvidia-ontology-management

✅ Overall verdict: PASS — Recommended for publication

Publication Recommendation

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

Evaluation Metadata

Skill: nvidia-ontology-management
Evaluation date: 2026-10-08
Evaluator version: 1.5.6
Agents: Claude Code (aws/anthropic/bedrock-claude-opus-4-8), Codex (openai/openai/gpt-5.5)
Tasks: 13 evaluation tasks (13 positive)
Dataset digest: sha256:4c8455d88fc45fd16ded90344c740ea7149fe29edbca7f112b346c21177d23fa (skill-evaluator-dataset-snapshot/1)
Attempts per task: 1
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	94.9% — baseline ran, but no comparable score was available; uplift unavailable	85.3% — baseline ran, but no comparable score was available; uplift unavailable
Security	100.0% → 100.0% (±0.0 points)	84.6% → 76.9% (-7.7 points)
Correctness	29.2% → 100.0% (+70.8 points)	47.7% → 89.2% (+41.5 points)
Discoverability	93.7% — baseline ran, but no comparable score was available; uplift unavailable	85.8% — baseline ran, but no comparable score was available; uplift unavailable
Effectiveness	42.4% → 89.3% (+46.9 points)	33.9% → 81.1% (+47.2 points)
Efficiency	91.4% — baseline ran, but no comparable score was available; uplift unavailable	93.3% — 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	2,136,288	2,211,810	-75,522	-3.41%	skill 13/13; base 13/13
claude-code	management-eval-001-mcp-readonly	176,205	177,943	-1,738	-0.98%	skill 1/1; base 1/1
claude-code	management-eval-002-validate-422	91,346	120,663	-29,317	-24.30%	skill 1/1; base 1/1
claude-code	management-eval-003-reset-is-destructive	97,849	150,421	-52,572	-34.95%	skill 1/1; base 1/1
claude-code	management-eval-004-token-acts-as-owner	95,573	119,914	-24,341	-20.30%	skill 1/1; base 1/1
claude-code	management-eval-005-import-replace	99,977	121,086	-21,109	-17.43%	skill 1/1; base 1/1
claude-code	management-eval-006-lineage-semantic-layer	285,753	119,687	+166,066	+138.75%	skill 1/1; base 1/1
claude-code	management-eval-007-patch-vs-put-sql-attribute	95,844	220,722	-124,878	-56.58%	skill 1/1; base 1/1
claude-code	management-eval-008-run-grain-measure	95,583	30,926	+64,657	+209.07%	skill 1/1; base 1/1
claude-code	management-eval-009-relationship-evidence	247,442	247,503	-61	-0.02%	skill 1/1; base 1/1
claude-code	management-eval-010-import-exact-readback	97,134	216,736	-119,602	-55.18%	skill 1/1; base 1/1
claude-code	management-eval-011-result-publication-mode	161,713	317,242	-155,529	-49.03%	skill 1/1; base 1/1
claude-code	management-eval-012-promotion-readback-failure	133,068	183,802	-50,734	-27.60%	skill 1/1; base 1/1
claude-code	management-eval-013-composed-lifecycle	458,801	185,165	+273,636	+147.78%	skill 1/1; base 1/1
codex	All cases	2,393,113	2,250,240	+142,873	+6.35%	skill 13/13; base 13/13
codex	management-eval-001-mcp-readonly	797,280	452,293	+344,987	+76.28%	skill 1/1; base 1/1
codex	management-eval-002-validate-422	57,538	54,886	+2,652	+4.83%	skill 1/1; base 1/1
codex	management-eval-003-reset-is-destructive	57,904	54,809	+3,095	+5.65%	skill 1/1; base 1/1
codex	management-eval-004-token-acts-as-owner	45,901	54,971	-9,070	-16.50%	skill 1/1; base 1/1
codex	management-eval-005-import-replace	161,338	55,064	+106,274	+193.00%	skill 1/1; base 1/1
codex	management-eval-006-lineage-semantic-layer	85,423	98,141	-12,718	-12.96%	skill 1/1; base 1/1
codex	management-eval-007-patch-vs-put-sql-attribute	45,034	65,171	-20,137	-30.90%	skill 1/1; base 1/1
codex	management-eval-008-run-grain-measure	44,493	54,668	-10,175	-18.61%	skill 1/1; base 1/1
codex	management-eval-009-relationship-evidence	413,138	83,561	+329,577	+394.41%	skill 1/1; base 1/1
codex	management-eval-010-import-exact-readback	44,648	55,235	-10,587	-19.17%	skill 1/1; base 1/1
codex	management-eval-011-result-publication-mode	118,995	475,882	-356,887	-74.99%	skill 1/1; base 1/1
codex	management-eval-012-promotion-readback-failure	110,830	262,196	-151,366	-57.73%	skill 1/1; base 1/1
codex	management-eval-013-composed-lifecycle	410,591	483,363	-72,772	-15.06%	skill 1/1; base 1/1
ALL AGENTS	Dataset aggregate	4,529,401	4,462,050	+67,351	+1.51%	skill 26/26; base 26/26

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); 5 finding(s)
Tier 2	Semantic deduplication	PASSED	2 validator(s); 0 finding(s)
Tier 3	Live agent evaluation	PASS	2 agent(s); 13 task(s)

Findings and Observations

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

MEDIUM QUALITY/quality_correctness: Resource-based skill lacks documentation of available resources (skills/nvidia-ontology-management/SKILL.md)
MEDIUM QUALITY/quality_reliability: MCP skill lacks connection/error guidance (skills/nvidia-ontology-management/SKILL.md)
MEDIUM QUALITY/quality_efficiency: Deeply nested references in modeling.md (skills/nvidia-ontology-management/SKILL.md)
LOW QUALITY/quality_reliability: Inputs are used but no dedicated Inputs section is documented (skills/nvidia-ontology-management/SKILL.md)
LOW QUALITY/quality_reliability: Structured output is used but no dedicated Output Format section is documented (skills/nvidia-ontology-management/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.

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

NVIDIA

Publisher

NVIDIA (Official)

Source

NVIDIA/skills · skills/nvidia-ontology-management · commit fce3cca

Licence

Apache License 2.0 (redistribution allowed)

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

Synced

2026-11-09

AutoGineer is not affiliated with or endorsed by NVIDIA or NVIDIA.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download nvidia-ontology-management.zip