# AI use cases for Open Vetta (openvetta/open-vetta)

Repo: [openvetta/open-vetta](https://github.com/openvetta/open-vetta) | Use cases: 5 | Page: https://dedupedai.com/use-cases/openvetta-open-vetta/

Open Vetta is an open-source, local-first desktop AI agent that works on your files, tools, and models (bring your own key) and can be extended with skills, MCP servers, plugins, and themes.

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

- Run a coding task on a local project with your own model: https://dedupedai.com/use-cases/openvetta-open-vetta/run-coding-task-on-local-project/
- Build and hot-reload a Vetta plugin outside the source repo: https://dedupedai.com/use-cases/openvetta-open-vetta/build-plugin-from-any-directory/
- Repeat a proven task across many directories on a schedule: https://dedupedai.com/use-cases/openvetta-open-vetta/scale-proven-task-across-directories/
- Build a local knowledge base for team reference: https://dedupedai.com/use-cases/openvetta-open-vetta/build-private-knowledge-base-for-team/
- Set up Open Vetta from source and run the desktop app: https://dedupedai.com/use-cases/openvetta-open-vetta/run-vetta-from-source-for-contributors/

## How to run a coding task on a local project with your own model with Open Vetta

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

Point Vetta at a local project and get code changes with visible tool calls and approvals.

**Who it is for:** A developer with a local repository who needs a bounded coding or file task done and wants to review every step the agent takes.

**The problem:** Chat tools that only return text leave the developer to copy changes by hand and cannot inspect the workspace or run commands in it.

**The outcome:** The agent inspects the project, makes changes or runs commands with permission prompts, and leaves the plan, tool calls, and results visible for review.

### Steps

1. Download the desktop app for your operating system from the Download link in the README, then follow the installation and first setup guide.
2. Add a model provider using BYOK: a supported provider, an OpenAI-compatible endpoint, or a local inference server.
3. Open your project folder as a workspace in a new session.
4. Follow the first-task walkthrough, or describe a small coding task such as fixing a failing test, and review each permission request before approving it.
5. Check the tool calls, plan, and results in the session history before accepting the changes.

### Prompt to paste into your agent

```text
Use the Open Vetta repo (openvetta/open-vetta) on the workspace at <project-path>. Ask Vetta to fix one small failing test or add one small function, approve only the commands you need, and report the list of files changed and the execution trail. Verify that the changed files match the task and that no unrequested files were modified.
```

### Caveat

Local-first does not mean no network traffic, because model providers, MCP servers, and plugins each create their own data boundary.

Risk, credentials: Your provider API key is kept in local credential storage, and model requests go to the endpoint you configure, so use a key with limited scope.

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/openvetta-open-vetta/run-coding-task-on-local-project/

## How to build and hot-reload a Vetta plugin outside the source repo with Open Vetta

Repo: [openvetta/open-vetta](https://github.com/openvetta/open-vetta) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README

Scaffold a plugin, install it into the running desktop app, and watch changes live.

**Who it is for:** A developer or coding agent that needs to add a custom desktop UI feature, file handler, message action, or tool to Vetta without cloning the Vetta source.

**The problem:** Extending a desktop agent usually means forking its source tree, which is slow to set up and hard for an agent to work inside.

**The outcome:** A working plugin project with an AGENTS.md file, built and installed into the desktop app, with hot reload during development.

### Steps

1. Run `npx @vetta-org/plugin-cli init --id my-plugin --name "My Plugin"` in an empty parent directory.
2. Run `cd my-plugin && npm install` to enter the project and install its dependencies.
3. Run `npx vetta-plugin-cli docs` to find the plugin manual and the SDK version it documents.
4. Run `npm run install:vetta` to build, package, and install the plugin into the running desktop app.
5. Run `npx vetta-plugin-cli watch` so the host loads the plugin from this directory and reloads on change.

### Prompt to paste into your agent

```text
Using the Open Vetta plugin CLI from the open-vetta project, scaffold a plugin named <plugin-id> in <parent-directory>, then add a small capability declared in plugin.json. Report the output of `npx vetta-plugin-cli docs` and confirm the plugin installs with `npm run install:vetta`. Verify that the declared capability appears in plugin.json and is limited to what the task requires.
```

### Caveat

The README says plugins should be treated as curated code, so read the plugin trust and permission model before distributing one.

Risk, remote script: Plugins run inside the desktop renderer and are treated as curated code, not an arbitrary-code sandbox, so review any plugin's code and permissions before installing it.

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/openvetta-open-vetta/build-plugin-from-any-directory/

## How to repeat a proven task across many directories on a schedule with Open Vetta

Repo: [openvetta/open-vetta](https://github.com/openvetta/open-vetta) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README

Run one working workflow over several folders with batch tasks, then schedule it as an automation.

**Who it is for:** A professional with a repeatable file or document task across many project folders, such as a team lead updating reports or docs in several repositories each week.

**The problem:** Running the same agent task by hand in each folder is slow and easy to get inconsistent.

**The outcome:** The same workflow runs across the chosen directories as a batch task, and can then be scheduled as an automation.

### Steps

1. Complete one task successfully in a single project by following the first-task walkthrough.
2. Save the working steps as a reusable skill or scenario from the abilities options in the product guide.
3. Create a batch task and select the directories that should receive the same workflow.
4. Review the execution trail for the first directory before letting the batch finish.
5. Turn the batch task into an automation if it should run on a repeating basis.

### Prompt to paste into your agent

```text
Use the Open Vetta repo (openvetta/open-vetta) to run a batch task built from the workflow in <workflow-name> across the directories in <directory-list>. Report, for each directory, whether the task finished and which files changed. Verify by checking that the results match the workflow on at least one directory you inspect by hand.
```

### Caveat

The README describes batch tasks and automation in general terms, so check the product guide for the exact setup before relying on a schedule for unattended work.

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/openvetta-open-vetta/scale-proven-task-across-directories/

## How to build a local knowledge base for team reference with Open Vetta

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

Put team documents into a local knowledge base the agent can use in later tasks.

**Who it is for:** A team lead or researcher who keeps internal procedures, notes, and reference files and wants an agent to answer from them without sending them to a hosted service.

**The problem:** Organizational knowledge is spread across files, and general chat models do not know the team's own procedures.

**The outcome:** A local knowledge base the agent can draw on, with the source files kept in the local workspace.

### Steps

1. Install the desktop app or run from source, then open a workspace for the team's documents.
2. Create a local knowledge base and add the reference files that the team uses most.
3. Ask the agent a question that depends on one of those files and check that its answer cites or matches the source.
4. Install a reusable skill or scenario for the team's recurring method if one applies.

### Prompt to paste into your agent

```text
Use the Open Vetta repo (openvetta/open-vetta) to create a local knowledge base from the files in <document-folder>. Ask it one question that only those files can answer and report the answer and the file it came from. Verify the answer against the source file by reading that file directly.
```

### Caveat

The README confirms local knowledge bases but does not describe how they index files, so test retrieval accuracy on your own documents before depending on it.

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/openvetta-open-vetta/build-private-knowledge-base-for-team/

## How to set up Open Vetta from source and run the desktop app with Open Vetta

Repo: [openvetta/open-vetta](https://github.com/openvetta/open-vetta) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README

Clone the repo, install dependencies, and start the desktop app in development mode.

**Who it is for:** A contributor or developer who wants to modify Vetta or test changes against the dev branch.

**The problem:** Contributing to a desktop agent needs a working development environment that does not touch the installed app's data.

**The outcome:** A running development build of the desktop app that uses its own data folder, with the repo ready for validation commands.

### Steps

1. Install Bun 1.3+ and Node.js 20+.
2. Run `git clone https://github.com/openvetta/open-vetta.git` and then `cd open-vetta`.
3. Run `git switch dev` and then `bun install`.
4. Run `cd apps/desktop` and then `bun run dev`.
5. Read QUICKSTART.md for the validation commands and run `bun run check:quick` before opening a pull request to the dev branch.

### Prompt to paste into your agent

```text
Clone openvetta/open-vetta, switch to the dev branch, run bun install, and start the desktop app from apps/desktop with bun run dev. Report whether the app launches and which data folder it uses. Verify that the data folder is ~/.vetta-dev and that the installed-app folder ~/.vetta was not changed.
```

### Caveat

Do not run bare bun test in this monorepo, and use the repository scripts instead, as the README instructs.

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/openvetta-open-vetta/run-vetta-from-source-for-contributors/