# AI use cases for Overtime (ukanwat/overtime)

Repo: [ukanwat/overtime](https://github.com/ukanwat/overtime) | Use cases: 5 | Page: https://dedupedai.com/use-cases/ukanwat-overtime/

Overtime is a TypeScript app that runs long-lived coding agents from a brief, letting them work on their own across sessions and message you only when a decision is yours.

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

- Keep CI and dependency PRs green while you work: https://dedupedai.com/use-cases/ukanwat-overtime/keep-ci-and-dependency-prs-green/
- Review each PR as it opens, worst findings first: https://dedupedai.com/use-cases/ukanwat-overtime/review-every-pr-as-it-opens/
- Wake an on-call agent the moment logs or metrics spike: https://dedupedai.com/use-cases/ukanwat-overtime/on-call-watch-that-wakes-on-alerts/
- Get a sourced research digest on a schedule: https://dedupedai.com/use-cases/ukanwat-overtime/scheduled-research-digest-with-sources/
- Close each month's spending from exported statements: https://dedupedai.com/use-cases/ukanwat-overtime/monthly-money-close-from-exports/

## How to keep CI and dependency PRs green while you work with Overtime

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

A repo-keeper agent watches CI and dependency PRs, fixes what breaks, and asks before anything ships.

**Who it is for:** A developer or maintainer whose repo gets failing CI runs and dependency update PRs between their own work sessions.

**The problem:** Broken CI and stale dependency PRs pile up while nobody is watching, and fixing them interrupts other work.

**The outcome:** An agent that keeps its own schedule, fixes breakages it can handle, and leaves a question with options and a recommendation before merging anything that ships.

### Steps

1. Run `npm install -g @ukanwat/overtime` to install the app.
2. Run `overtime` and type a name for the agent, for example repo-keeper.
3. Check the backend, model, daily budget and folder, then press Enter.
4. Brief it in plain words: watch CI and dependency PRs in your repo, fix what breaks, and ask before merging anything that ships.
5. Set any paths you want protected as read-only so the agent cannot change them.

### Prompt to paste into your agent

```text
Using the Overtime repo at ukanwat/overtime, set up an agent named repo-keeper briefed to watch CI and dependency PRs for <target repo>. Expected output: a short list of failing checks and open dependency PRs, with a fix or a question for each. Verify that no merge happens without an approval question being answered.
```

### Caveat

Overtime is in early development and Claude is the backend tested most, so Codex and Gemini CLI paths need wider testing.

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/ukanwat-overtime/keep-ci-and-dependency-prs-green/

## How to review each PR as it opens, worst findings first with Overtime

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

A reviewer agent reads each new PR, splits large ones across helpers, and lists findings worst first.

**Who it is for:** A team lead or solo maintainer who wants a first-pass review on every new pull request without waiting for a human reviewer.

**The problem:** PRs wait hours or days for a first review, and large changes are hard to review in one pass.

**The outcome:** A reviewer agent that wakes when a PR opens and leaves findings ordered from most to least serious, with large diffs split across helper agents.

### Steps

1. Run `npm install -g @ukanwat/overtime` to install the app.
2. Run `overtime` and name the agent reviewer.
3. Check the backend and budget settings, then press Enter.
4. Brief it: review each PR as it opens, split big ones across helpers, and leave findings worst first.
5. Open a test PR and read the findings the agent leaves.

### Prompt to paste into your agent

```text
Using the Overtime repo at ukanwat/overtime, create a reviewer agent that reviews each new PR in <target repo>. Expected output: a findings list ordered worst first for one open PR. Verify that large diffs were split across helpers and that every finding names a file.
```

### Caveat

The README does not describe how the reviewer connects to your git host, so confirm the access setup before relying 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/ukanwat-overtime/review-every-pr-as-it-opens/

## How to wake an on-call agent the moment logs or metrics spike with Overtime

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

A watch on your logs or metrics wakes an agent within seconds to find the cause and fix or roll back.

**Who it is for:** An engineer on call for a service who wants a first responder that is awake even when they are not.

**The problem:** Incidents start at night or during other work, and the first minutes go to finding which change caused the spike.

**The outcome:** An agent that wakes on a watch, investigates the cause, rolls back or fixes it, and writes a postmortem for you to review.

### Steps

1. Run `npm install -g @ukanwat/overtime` to install the app.
2. Run `overtime` and name the agent on-call.
3. Brief it with the logs or metrics it should watch and what it may do: find the cause, roll back or fix, and write a postmortem.
4. Set up a watch on your log or metric source as described in the how-it-works guide, using the watch script the agent creates. (inferred, check the README)
5. Trigger a test event and check that the agent wakes and writes a postmortem.

### Prompt to paste into your agent

```text
Using the Overtime repo at ukanwat/overtime, create an on-call agent with a watch on <log or metric source>. Expected output: after a test alert, a cause summary, the action taken, and a postmortem file. Verify that the agent wakes within the watch interval and asks before any change it has not been permitted to make.
```

### Caveat

The README describes the watch as a small script that wakes the agent but does not list which log or metric sources are supported, so check that yours can be watched.

Risk, credentials: The agent needs access to your logs or metrics and possibly your deploy tooling, so grant only the scopes it needs and protect paths it must not change.

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/ukanwat-overtime/on-call-watch-that-wakes-on-alerts/

## How to get a sourced research digest on a schedule with Overtime

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

A research agent reads papers, docs or competitors on a schedule and reports only when something matters.

**Who it is for:** A product, research or strategy person who needs to track a field, docs or competitors without reading every update.

**The problem:** Keeping up with papers, documentation changes and competitor moves takes hours each week and is easy to skip.

**The outcome:** A research agent that runs on its own schedule, reads what you point it at, and reports with sources only when something is worth knowing.

### Steps

1. Run `npm install -g @ukanwat/overtime` to install the app.
2. Run `overtime` and name the agent research.
3. Brief it: which papers, docs or competitors to read, how often, and that it should report with sources only when something is worth knowing.
4. Check the first report it writes for sources and relevance.
5. Adjust the brief in its AGENT.md file if the reports are too frequent or too thin.

### Prompt to paste into your agent

```text
Using the Overtime repo at ukanwat/overtime, create a research agent that reads <list of sources> on a schedule. Expected output: one report with a source link for each item. Verify that every claim in the report links to a source the agent actually read.
```

### Caveat

Overtime is early and the README does not say how sources are fetched, so check that each source is reachable from your setup before relying on the digest.

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/ukanwat-overtime/scheduled-research-digest-with-sources/

## How to close each month's spending from exported statements with Overtime

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

A money agent reads the statements you export, checks spending against budget, and flags unused subscriptions.

**Who it is for:** A household or freelancer who exports statements each month and wants the spending review done without a spreadsheet session.

**The problem:** Monthly spending reviews are skipped, so overspending and forgotten subscriptions go unnoticed for months.

**The outcome:** A monthly close showing spending against your budget and a list of subscriptions you no longer use, based only on the files you export.

### Steps

1. Run `npm install -g @ukanwat/overtime` to install the app.
2. Run `overtime` and name the agent money.
3. Brief it: each month, close the month from the statements I export, compare spending to my budget, and list subscriptions I no longer use.
4. Place your exported statement files in the agent folder or the folder you point it to.
5. Read the first close and confirm the totals match your statements.

### Prompt to paste into your agent

```text
Using the Overtime repo at ukanwat/overtime, create a money agent that closes the month from the statement exports in <folder>. Expected output: spending by category against budget and a list of possibly unused subscriptions. Verify that the totals match the sum of the exported lines.
```

### Caveat

The README says the close works from exports you provide, so it does not connect to your bank and you must export statements yourself.

Risk, credentials: Bank and card statements are sensitive, so keep exports in a protected folder and avoid pasting account numbers into the brief.

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/ukanwat-overtime/monthly-money-close-from-exports/