# AI use cases for Agentbaseline (agentbaseline/agentbaseline)

Repo: [agentbaseline/agentbaseline](https://github.com/agentbaseline/agentbaseline) | Use cases: 3 | Page: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/

Agent Baseline is a draft framework of six security outcomes and 35 controls an enterprise must evidence to run AI agents safely, kept as machine-readable YAML.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Check your AI agent program against 35 controls: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/gap-assess-agent-program-against-controls/
- Load the controls into your GRC tooling: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/parse-controls-yaml-for-tooling/
- Regenerate the white paper figures from the YAML: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/rebuild-baseline-figures/

## How to check your AI agent program against 35 controls with Agentbaseline

Repo: [agentbaseline/agentbaseline](https://github.com/agentbaseline/agentbaseline) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Use the six outcomes and 35 controls to find gaps in how your company runs agents.

**Who it is for:** A CISO asked whether the company is ready to deploy agents widely.

**The problem:** Agent risk does not map neatly to product categories.

**The outcome:** A gap list against outcome-based controls.

### Steps

1. Read the six outcomes starting with Discover
2. Open whitepaper/controls.yaml
3. Mark each control as met, partial or missing
4. Prioritize missing controls

### Prompt to paste into your agent

```text
Using agentbaseline/agentbaseline whitepaper/controls.yaml, build a gap table for <organization> with each control id, title and status. Cite control ids exactly.
```

### Caveat

It is v1.0-draft, not a standard, and identifiers are not frozen until v1.0.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/gap-assess-agent-program-against-controls/

## How to load the controls into your GRC tooling with Agentbaseline

Repo: [agentbaseline/agentbaseline](https://github.com/agentbaseline/agentbaseline) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Parse the YAML source of truth to import outcomes and controls into your own systems.

**Who it is for:** A GRC engineer automating control tracking.

**The problem:** Copying controls from a PDF is error-prone.

**The outcome:** Controls imported with ids, outcomes, types and status.

### Steps

1. Clone the repo
2. Read whitepaper/controls.yaml
3. Map id, outcome, type and status fields to your tool
4. Re-import when versions change

### Prompt to paste into your agent

```text
Parse whitepaper/controls.yaml from agentbaseline and output CSV rows of id, outcome, title, type and status. Confirm 35 controls were parsed.
```

### Caveat

Check VERSIONING.md before citing identifiers externally.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/parse-controls-yaml-for-tooling/

## How to regenerate the white paper figures from the YAML with Agentbaseline

Repo: [agentbaseline/agentbaseline](https://github.com/agentbaseline/agentbaseline) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README

Run the figure scripts so diagrams stay in sync with control changes.

**Who it is for:** A contributor proposing changes to the baseline.

**The problem:** Figures drift from the source data.

**The outcome:** Figures rebuilt from controls.yaml.

### Steps

1. Clone the repo
2. Run `python3 whitepaper/figures/build-figure-1.py`
3. Run `python3 software-factory/figures/build-figures.py`
4. Review the outputs

### Prompt to paste into your agent

```text
Run the agentbaseline figure scripts and confirm Figure 1 regenerates without errors.
```

### Caveat

The public comment period closed on 30 September 2026.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/agentbaseline-agentbaseline/rebuild-baseline-figures/