---
title: "Undominated · Model card transparency audit: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/skills/undominated-model-card-transparency-audit/
description: "Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data, compute hours, and limitations."
---

# Undominated · Model card transparency audit: review, setup & limitations · Undominated.ai

> Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data, compute hours, and limitations.

[← Explore all skills](/skills/)

EVIDENCE AND EVALUATION / Undominated.ai

# Undominated · Model card transparency audit

Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data, compute hours, and limitations.

 [See setup guidance ↓](#setup)[Original source ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md)

SOURCE REVIEW

 Reviewed 2026-10-14
 Evidence 3 linked sources
 Publisher Undominated.ai
 Licence [MIT ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/LICENSE.txt)
 Revision 30751bb01d24
 [Read what was—and wasn’t—checked ↓](#review)

“Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data, compute hours, and limitations.”

 [Undominated.ai · upstream description ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - Pairs a bounded workflow with an offline Python checker and explicitly synthetic example inputs.
- Created for recurring evidence and comparison failures: missing facts remain unknown and every conclusion keeps its scope.
- It names a concrete failure mode: Flag critical opacity gaps (e.g. models hiding training cut-off dates or omitting safety evaluations).

### A good fit for

 - Score foundation model documentation transparency across data provenance, compute disclosure, carbon footprint, and evaluation limits.
- Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data sources.

### Weigh up before choosing

 - Checks presence of documented transparency indicators; does not verify independent accuracy of training data logs.
- Fixture checks establish the documented local behaviour, not performance on an unrestricted set of real tasks.
- Public GitHub source availability is not yet verified; the complete download contains the reviewed source.
- A new npm install path is not yet verified. Use the complete source download.
- `0`: Valid audit meeting all transparency thresholds.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - skills/undominated-model-card-transparency-audit/SKILL.md
- skills/undominated-model-card-transparency-audit/scripts/check.py
- skills/undominated-model-card-transparency-audit/examples/synthetic.json

### Our findings

 - First-party resource authored by Undominated.ai; this is our own verification record, not an independent endorsement.
- The download preserves the complete source and supporting files. Its manifest hashes identify the exact bytes.
- One input rule the file states is: `minTransparencyScorePercent`: optional decimal string (e.g. "75.0").
- One condition that blocks a pass is: model and publisher must be non-empty strings.

### Not established by this review

 - Model compliance with these instructions on arbitrary tasks
- Native integration in every agent host
- Public GitHub source identity
- Remote installation through the standard Skills CLI
- The new npm resource CLI installation

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Add a skill to your workflow.

[Upstream setup instructions ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md)
 - Download the complete bundle from https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/bundle.zip and extract it into a new directory.
- Place the extracted directory at .agents/skills/undominated-model-card-transparency-audit/ or your host's documented skills directory. Preserve SKILL.md, scripts, examples and licence files together.
- Run the documented synthetic example from the skill directory with Python 3.10 or later before using your own evidence.

### Before you start

 - An Agent Skills compatible host
- Python 3.10 or later for the optional checker

### Compatibility

Agent Skills compatible hosts · Python 3.10+

### Implementation

Python · Markdown

### Local, explicit inputs

 - read:user-selected-local-file

### Cost model

MIT source at no charge. Your agent host or model provider may charge for use; the included offline checks require no paid API.

THE COMPLETE REVIEWED DEFINITION

## Read it before you reuse it.

Original source bytes, with attribution. Review the host-specific setup notes above.

 Copy definition ↗ [Download complete resource + licence ↗](/resources/skills/undominated-model-card-transparency-audit/bundle.zip)[Raw Markdown ↗](/resources/skills/undominated-model-card-transparency-audit/definition.md)
 # Model Card Transparency Audit

Audit model card releases against the Foundation Model Transparency Index, checking documentation completeness across pre-training data sources, compute hours, hardware types, carbon disclosures, and safety limitations.

## Workflow

1. Ingest model documentation audit checklist across 5 core transparency dimensions:
 - Data provenance (data sources disclosed, filtering method described, copyright handling).
 - Compute infrastructure (hardware type, cluster size, total FLOPs or GPU-hours disclosed).
 - Environmental impact (energy consumption in MWh, carbon emissions disclosed).
 - Model capabilities & limits (intended use cases, out-of-scope uses, known failure modes).
 - Evaluation methodology (benchmark harnesses named, prompts published, contamination checked).
2. Calculate transparency disclosure score:
 $$\text{Transparency Score} = \frac{\sum \text{Disclosed Indicators}}{\text{Total Indicators}} \times 100\%$$
3. Flag critical opacity gaps (e.g. models hiding training cut-off dates or omitting safety evaluations).

## Synthetic Input Format

JSON object with:
- `schemaVersion`: 1
- `model`: string
- `publisher`: string
- `disclosures`:
 - `dataSourcesNamed`: boolean
 - `dataFilteringDescribed`: boolean
 - `hardwareArchitectureDisclosed`: boolean
 - `computeBudgetDisclosed`: boolean
 - `carbonEmissionsReported`: boolean
 - `knownLimitationsDocumented`: boolean
 - `evaluationHarnessCodePublic`: boolean
 - `safetyRedTeamingReported`: boolean
- `assertions`:
 - `minTransparencyScorePercent`: optional decimal string (e.g. "75.0")
 - `requireLimitationsDocumented`: optional boolean (default: true)

## Exit Codes

- `0`: Valid audit meeting all transparency thresholds.
- `1`: Transparency score below requirement or mandatory disclosure missing.
- `2`: Malformed input or safety violation.

## Verification

```sh
python3 scripts/check.py examples/synthetic.json
```

The download contains SKILL.md . Extract the whole bundle; the supporting files are required. Inspect the included MANIFEST.json for file hashes.

By **Undominated.ai**. [Exact upstream source ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md) · [Licence](/resources/skills/undominated-model-card-transparency-audit/LICENSE.txt) · [Attribution](/resources/skills/undominated-model-card-transparency-audit/ATTRIBUTION.txt)

SHA-256 30751bb01d247d2b9198a79a3b9312bed91bfb2219333c37fd8f6b906e7f75a0

 Read the applicable licence MIT License

Copyright (c) 2026 Undominated.ai

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [Original Undominated definition ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md) Checked 2026-10-07 https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/definition.md Supports: Workflow and input requirements, Stated limitations
- [MIT licence ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/LICENSE.txt) Checked 2026-10-07 https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/LICENSE.txt Supports: Redistribution terms and attribution
- [Complete source bundle ↗](https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/bundle.zip) Checked 2026-10-07 https://undominated.ai/resources/skills/undominated-model-card-transparency-audit/bundle.zip Supports: Full local source, supporting files and integrity manifest

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Undominated · EU AI Act Transparency Audit ↗ Checks four supplier declarations in a local JSON file: training compute, copyright summary, energy log, and model card. A true declaration is not an EU AI Act determination.](/skills/undominated-eu-ai-act-transparency-audit/)[### Undominated · Token blowout circuit breaker audit ↗ Audits runaway loop detection and trips circuit breaker when token expenditures exceed budget ceilings.](/skills/undominated-token-blowout-circuit-breaker-audit/)[### Undominated · Zero Data Retention policy audit ↗ Audit provider Zero Data Retention (ZDR) claims against published Terms of Service, abuse monitoring windows, logging exceptions, and training opt-out guarantees.](/skills/undominated-zdr-policy-audit/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Choose a toolkit](/tools/)
- [Inspect execution hosts](/agents/)
- [Check tool access](/mcp-servers/)
- [Check project dependencies](/check-repo/)
