# AI use cases for Pilot Deck (OpenBMB/PilotDeck)

Repo: [OpenBMB/PilotDeck](https://github.com/OpenBMB/PilotDeck) | Use cases: 3 | Page: https://dedupedai.com/use-cases/openbmb-pilotdeck/

PilotDeck is an open-source, WorkSpace-based agent operating system from OpenBMB and partners for long-running tasks, with visual configuration, many model providers including Ollama, and Word, Excel and PPT skills.

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

- Install a self-hosted agent workspace in one command: https://dedupedai.com/use-cases/openbmb-pilotdeck/one-click-install-agent-workspace/
- Run the agent platform fully on a local model: https://dedupedai.com/use-cases/openbmb-pilotdeck/run-agent-platform-on-local-ollama/
- Produce Word, Excel and PPT files with an agent: https://dedupedai.com/use-cases/openbmb-pilotdeck/generate-office-docs-with-agent/

## How to install a self-hosted agent workspace in one command with Pilot Deck

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

Install Pilot Deck with its one-line script, set up a model provider visually in the web UI, and run long tasks inside isolated WorkSpaces.

**Who it is for:** A builder or small team that wants a self-hosted agent platform for long-running work, with each project kept in its own workspace.

**The problem:** General chat agents mix projects together and lose track on long tasks, and hosted agent platforms keep your files on their servers.

**The outcome:** Pilot Deck running at localhost:3001 with your provider configured, WorkSpaces per project, and tasks that stay on track over long sessions.

### Steps

1. On macOS or Linux run `curl -fsSL https://raw.githubusercontent.com/OpenBMB/PilotDeck/main/install.sh | bash`
2. Start it with `pilotdeck`, which serves the app at http://localhost:3001
3. Open the URL; with no config file it opens onboarding where you add a provider, API key and model
4. Saving a valid provider writes ~/.pilotdeck/pilotdeck.yaml and starts the Gateway
5. Create a WorkSpace for one project and give it a long task
6. Check health any time with `pilotdeck status`

### Prompt to paste into your agent

```text
Install OpenBMB/PilotDeck on this machine with the install script, start it, and tell me the URL. Then explain what goes in ~/.pilotdeck/pilotdeck.yaml for <my provider>, without writing my API key into any file you show me.
```

### Caveat

Needs Node.js 22.13 or newer and below 23; the script checks this. Results depend on the model you choose.

Risk, remote script: The installer pipes a script from GitHub into bash; read it first.

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/openbmb-pilotdeck/one-click-install-agent-workspace/

## How to run the agent platform fully on a local model with Pilot Deck

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

Point Pilot Deck at Ollama through its OpenAI compatible protocol so the agent runs on your own machine with no cloud API key.

**Who it is for:** A privacy-minded developer or tinkerer who wants to try an agent workspace without sending prompts or files to a cloud model.

**The problem:** Agent platforms usually assume a paid cloud API, which costs money and sends your data off the machine.

**The outcome:** Pilot Deck using a model served by Ollama at localhost:11434, configured in one YAML file, for small local tasks.

### Steps

1. Install Pilot Deck and Ollama
2. Start Ollama with `ollama serve`
3. Pull a model: `ollama pull qwen3:0.6b`, or a larger one your hardware can run
4. In ~/.pilotdeck/pilotdeck.yaml set agent model to ollama/qwen3:0.6b and add an ollama provider with protocol openai and url http://localhost:11434/v1
5. Restart Pilot Deck with `pilotdeck` and open http://localhost:3001
6. Run a small task in a new WorkSpace and check the replies come from the local model

### Prompt to paste into your agent

```text
Configure OpenBMB/PilotDeck to use Ollama with <model>, following the README's ollama YAML example. Restart it, run a short test task, and tell me whether the model is fast enough on this machine.
```

### Caveat

qwen3:0.6b is tiny; long or complex tasks need a much larger model.

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/openbmb-pilotdeck/run-agent-platform-on-local-ollama/

## How to produce Word, Excel and PPT files with an agent with Pilot Deck

Repo: [OpenBMB/PilotDeck](https://github.com/OpenBMB/PilotDeck) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Use Pilot Deck 2.0's Word, Excel and PPT skills inside a WorkSpace to turn notes and data into office-ready documents.

**Who it is for:** An analyst, manager or consultant who spends hours turning notes into slide decks, spreadsheets and reports.

**The problem:** Formatting decks and spreadsheets by hand is slow, and chat agents return text you still have to paste into Office.

**The outcome:** Real .pptx, .xlsx and .docx files built from your WorkSpace material, with layout and tone kept separate per project.

### Steps

1. Install Pilot Deck with the one-line script and start it with `pilotdeck`
2. Configure a capable model in onboarding or in pilotdeck.yaml
3. Create a WorkSpace for the project and add your notes, data or past documents
4. Ask for the output, for example a 10 slide PPT deck and an Excel summary table
5. Open the files and ask for changes in the same WorkSpace so memory stays with that project
6. Keep separate WorkSpaces for clients so formats and tone do not bleed across

### Prompt to paste into your agent

```text
In a new PilotDeck WorkSpace called <project>, use the PPT skill to build a <n> slide deck from these notes: <notes>, and the Excel skill to build a summary table of <data>. List the files created.
```

### Caveat

Quality depends on the model you configure. Building from source needs GIT_LFS_SKIP_SMUDGE and exact Node version limits.

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/openbmb-pilotdeck/generate-office-docs-with-agent/