# AI use cases for Agentacct (mikehasa/agentacct)

Repo: [mikehasa/agentacct](https://github.com/mikehasa/agentacct) | Use cases: 4 | Page: https://dedupedai.com/use-cases/mikehasa-agentacct/

agentacct is a local-first dashboard and terminal app that shows what your coding agents did, the tokens and estimated cost behind that work, and whether their claimed results were verified.

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

- See what your coding agents did and what it cost: https://dedupedai.com/use-cases/mikehasa-agentacct/see-what-coding-agents-did-and-cost/
- Check whether an agent's 'done' claim is verified: https://dedupedai.com/use-cases/mikehasa-agentacct/check-whether-an-agent-done-claim-holds/
- Review token use and cost by agent and model: https://dedupedai.com/use-cases/mikehasa-agentacct/review-token-usage-and-cost-by-agent-and-model/
- Follow one project folder across every agent: https://dedupedai.com/use-cases/mikehasa-agentacct/follow-one-project-across-every-agent/

## How to see what your coding agents did and what it cost with Agentacct

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

Install Agentacct once and review recent agent tasks, tool calls, tokens, and estimated cost on your machine.

**Who it is for:** A developer who runs several coding agents across projects and wants to know what they worked on last week and roughly what it cost.

**The problem:** Agent sessions are spread across client log files, so it is hard to tell which task changed which files, what ran, and how many tokens it used.

**The outcome:** A local dashboard and terminal app listing recent tasks with their receipts, steps, checks, tokens, and cost estimates, with nothing sent off the machine.

### Steps

1. Run `pipx install agentacct`
2. Run `agentacct onboard` once on this machine so it finds your agents and starts the recorder
3. Open a new agent session so hooks and MCP servers bind
4. Run `agentacct tui` and open Sessions to see the most recently active tasks

### Prompt to paste into your agent

```text
Install agentacct from https://github.com/mikehasa/agentacct on this machine using pipx, then run agentacct onboard. Report the most recent task names and their estimated costs in <project path>, and confirm that costs are marked as estimates. Verify that the Sessions view opens in agentacct tui.
```

### Caveat

Costs are pricing-table estimates rather than invoices, and Windows is supported only through WSL.

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/mikehasa-agentacct/see-what-coding-agents-did-and-cost/

## How to check whether an agent's 'done' claim is verified with Agentacct

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

Open a task receipt and separate the agent's own Reported result from checks that passed after the last change.

**Who it is for:** A developer who has been told an agent finished a change and wants to confirm it before merging or reviewing the diff.

**The problem:** An agent's own statement that work is done can sound final even when no passing check ran after its last edit.

**The outcome:** A clear view of whether the result is only Reported or Verified, with the timeline of attempts, checks, and any earlier failures still visible.

### Steps

1. Run `agentacct tui`
2. Open Sessions and select the task whose result you want to check
3. Read the recorded outcome: Reported means the agent's own claim, Verified means current passing checks after the last recorded change
4. Open the check details and the activity timeline to see which tests ran and when

### Prompt to paste into your agent

```text
Use the agentacct repo at https://github.com/mikehasa/agentacct to open the receipt for the task <task name> and report whether its result is Reported or Verified. Confirm that any passing checks ran after the last recorded change, and list any earlier failed checks that remain in the history.
```

### Caveat

A Verified result only covers the checks Agentacct recorded, so it does not replace your own review of the code.

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/mikehasa-agentacct/check-whether-an-agent-done-claim-holds/

## How to review token use and cost by agent and model with Agentacct

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

Filter usage by date, agent, model, or provider to see where tokens and estimated spend went.

**Who it is for:** A developer or team lead who pays for coding-agent usage and wants to compare agents and models over a set period.

**The problem:** Usage is split across each client's own records, so comparing agents, models, and periods takes manual work.

**The outcome:** A usage page that totals tokens and estimated cost by agent and model for a chosen range, with cache tokens shown separately and estimates labeled.

### Steps

1. Run `agentacct tui` and open the usage view
2. Set the date range, then filter by agent, model, or provider
3. Read the totals by agent and by model for the selected range
4. Check the freshness stamp in the top bar before trusting the numbers

### Prompt to paste into your agent

```text
Use the agentacct repo at https://github.com/mikehasa/agentacct to report token totals and estimated cost by agent and by model for the last 7 days on this machine. Verify that cost figures are labeled as estimates and that the local data freshness stamp is shown.
```

### Caveat

Dates show session totals by activity date, so sessions that span several days are not split into exact daily spend.

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/mikehasa-agentacct/review-token-usage-and-cost-by-agent-and-model/

## How to follow one project folder across every agent with Agentacct

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

Point a work group at a project folder to see which agents worked there and in what order.

**Who it is for:** A developer who switches between several coding agents on the same repository and loses track of which one changed what.

**The problem:** Work on one project is scattered across sessions from different agents, so no single view shows the sequence of changes.

**The outcome:** A folder-anchored Work group with its sessions drawn on one cross-agent timeline, where you can open and scrub each session.

### Steps

1. Run `agentacct tui`
2. Open the Work tab
3. Select the group for your project folder
4. Open a session from the shared timeline with the enter key and scrub through its steps

### Prompt to paste into your agent

```text
Using the agentacct repo at https://github.com/mikehasa/agentacct, open the Work group for the folder <project path> and list the agents that worked there and their most recent sessions. Verify that group totals are shown as sums of the participating sessions, with partial coverage labeled.
```

### Caveat

Each session keeps its own evidence, and the agent that worked on a folder only appears when its logs are readable on this machine.

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/mikehasa-agentacct/follow-one-project-across-every-agent/