# AI use cases for Errand (runta-dev/errand) Repo: [runta-dev/errand](https://github.com/runta-dev/errand) | Use cases: 4 | Page: https://dedupedai.com/use-cases/runta-dev-errand/ Errand is an open-source macOS desktop app built with Electron and TypeScript, where you create persistent AI agents that each run on their own Runta cloud computer. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run a persistent AI agent on a focused job: https://dedupedai.com/use-cases/runta-dev-errand/run-persistent-agent-on-focused-job/ - Send a screenshot to an agent as a follow-up: https://dedupedai.com/use-cases/runta-dev-errand/send-image-with-agent-follow-up/ - Fix a bug in Errand and run its checks before a pull request: https://dedupedai.com/use-cases/runta-dev-errand/contribute-fix-to-errand/ - Build a local macOS package and smoke test it: https://dedupedai.com/use-cases/runta-dev-errand/package-and-smoke-test-macos-build/ ## How to run a persistent AI agent on a focused job with Errand Repo: [runta-dev/errand](https://github.com/runta-dev/errand) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Give an agent its own cloud computer and one job, then follow its work from your Mac. **Who it is for:** A Mac user with a recurring or multi-step task who wants an AI teammate that keeps its files and history between sessions. **The problem:** Chat tools forget context and give you no workspace, so every follow-up means rebuilding the situation from scratch. **The outcome:** An agent with its own cloud workspace is working on your job, and you can reopen the app later to see its conversation, tool activity, and latest results. ### Steps 1. Run `git clone https://github.com/runta-dev/errand.git` 2. Run `cd errand` then `npm ci` 3. Run `npm run dev` and sign in through your browser with your Runta account 4. Create an agent and give it a focused job 5. Follow the work in the conversation, tool activity, or live desktop, then reopen Errand later to pick up the latest results ### Prompt to paste into your agent ```text Using the runta-dev/errand repository, install dependencies with npm ci and start the app with npm run dev, then create an agent with the job . Expected output: a running agent whose conversation shows tool activity for that job. Verify that the conversation is still present after you close and reopen Errand. ``` ### Caveat The app connects to Runta Cloud Agents by default, so it does nothing useful without a Runta account and configured model access. 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/runta-dev-errand/run-persistent-agent-on-focused-job/ ## How to send a screenshot to an agent as a follow-up with Errand Repo: [runta-dev/errand](https://github.com/runta-dev/errand) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Attach an image to a message so the agent sees it as native image input in the same conversation. **Who it is for:** A user who is already working with an agent and needs to show it a visual problem, such as a layout, chart, or error screen. **The problem:** Describing a visual issue in text takes effort and loses detail, and starting a new chat loses the earlier context. **The outcome:** The agent receives the image as native input inside the existing conversation, and the attachment stays with that conversation for later reference. ### Steps 1. Open an existing agent conversation in Errand 2. Attach an image to your message 3. Send the follow-up message with a specific question about the image 4. Check the agent's reply and the tool activity for how it used the image 5. Reopen the conversation later to confirm the attachment is still there ### Prompt to paste into your agent ```text In the runta-dev/errand app, open the conversation for the agent and attach the image to a follow-up asking . Expected output: the agent's reply refers to content from the image. Verify that the attachment is still shown in the conversation after reopening Errand. ``` ### Caveat This depends on an existing agent and a signed-in Runta account, and the README does not list which image formats are accepted. 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/runta-dev-errand/send-image-with-agent-follow-up/ ## How to fix a bug in Errand and run its checks before a pull request with Errand Repo: [runta-dev/errand](https://github.com/runta-dev/errand) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Clone the repo, run the typecheck, lint, test, and build scripts, then submit a fix. **Who it is for:** A developer who has hit a rough edge in Errand, such as a confusing activity state or a connection problem, and wants to fix it. **The problem:** Without a reproducible report and passing checks, a fix is hard for maintainers to review and may break the build. **The outcome:** A tested change is ready as a pull request, with steps to reproduce and the expected behavior written up as the project asks. ### Steps 1. Run `git clone https://github.com/runta-dev/errand.git` and `cd errand` 2. Run `npm ci` and `npm run dev` to reproduce the issue 3. Make your change, then run `npm run typecheck`, `npm run lint`, `npm test`, and `npm run build` 4. Open a pull request at https://github.com/runta-dev/errand/pulls that includes the steps to reproduce, what you expected, and what happened ### Prompt to paste into your agent ```text In the runta-dev/errand repository, reproduce the bug using npm ci and npm run dev, then apply a fix. Expected output: a diff where npm run typecheck, npm run lint, npm test, and npm run build all pass. Verify each script exits with success before you write the pull request description. ``` ### Caveat The README asks for reproducible reports, so an unverified guess about a bug is less likely to be accepted than a fix with clear reproduction steps. 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/runta-dev-errand/contribute-fix-to-errand/ ## How to build a local macOS package and smoke test it with Errand Repo: [runta-dev/errand](https://github.com/runta-dev/errand) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Produce ad-hoc signed Errand builds in release/ and run the smoke check for testing. **Who it is for:** A developer or tester who changed the desktop client and needs a local installable build to check before a release. **The problem:** Testing packaging only through a public release is slow, and the signed release path is not meant for local experiments. **The outcome:** Local macOS artifacts are written to release/ and the smoke script reports on the packaged build. ### Steps 1. Run `git clone https://github.com/runta-dev/errand.git` and `cd errand`, then run `npm ci` 2. Run `npm run package` 3. Run `npm run smoke` 4. Open the artifacts in the `release/` folder and confirm the DMG and ZIP are present ### Prompt to paste into your agent ```text In the runta-dev/errand repository, run npm ci, then npm run package and npm run smoke. Expected output: artifacts in the release/ directory and a passing smoke run. Verify that the DMG contents and blockmap are present before reporting success. ``` ### Caveat Local builds are ad-hoc signed for testing only, so the README says distribution must go through the signed release workflow. 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/runta-dev-errand/package-and-smoke-test-macos-build/