# dira vs. Clockify, Toggl and the stopwatch

> Time trackers were built for one human at one keyboard — you run three agents.

Clockify, Toggl and the timer-plus-JIRA-worklog routine all measure the same
thing: a person, manually, after the fact. None of them can see the AI coding
agents doing the work — or prove a single minute of it.

Canonical page: https://dirahq.sh/compare

## You already log time twice. Neither place is right.

A timer you forget to start, and a JIRA worklog you fill from memory at 6pm.
Two reconstructions of a day that three agents actually built.

**A normal Tuesday, today**

| Time | What happens |
| --- | --- |
| 09:12 | Start a timer in Toggl. Forget which task. |
| 09:40 | Kick off two Claude Code sessions and a Codex run. |
| 11:05 | Timer still running through lunch. Was that billable? |
| 14:20 | Switch repos. Forget to switch the timer. |
| 18:00 | Reconstruct it all into JIRA worklogs from memory. |

Double entry, guessed from memory, agents invisible.

**The same Tuesday, with dira**

| Time | What happens |
| --- | --- |
| 09:12 | You just start working. Nothing to press. |
| 09:40 | Three agent sessions captured automatically, per repo. |
| 11:05 | Lunch idles out after 5 minutes — not counted. |
| 14:20 | Repo change detected — time re-attributes itself. |
| 18:00 | Go home. The invoice already has the commits behind it. |

Zero entry, deduped and idle-trimmed, git-anchored.

## Feature by feature

| Capability | Manual + JIRA | Clockify / Toggl | dira |
| --- | --- | --- | --- |
| How a minute is recorded | Manual, twice — timer + worklog | Manual, start/stop a timer | Automatic, agent hooks, no input |
| Running several agents at once | Blind — agents are invisible | Blind — one timer, one stream | Deduped to one human-minute |
| AI tokens & compute cost | No concept of it | No concept of it | Reconciled per session |
| Proof behind the number | Self-reported | Self-reported | Git-anchored commits |
| Where you do it | 2 apps — browser + ticket | 1 app, separate window | In the terminal you're already in |
| End-of-day reconstruction | Daily, rebuilt from memory | Often, fixing forgotten timers | Never — captured as it happens |
| What you bill on | Flat hours, one number | Flat hours, one number | Split: engaged / agent / compute |
| Rate change re-prices old invoices | Manual rework | Manual rework | Date-stamped policy |

Clockify and Toggl are excellent manual timers. They were simply built before
agents wrote the code.

## Three things only an agent-native tracker knows

### Parallel agents, deduped

A stopwatch records one stream. Supervise three agents and a manual timer
either triple-counts or sees nothing. dira counts one human-minute per minute
and splits it across the repos that earned it.

### Compute, as its own line

Toggl has no concept of a token. dira reconciles real usage from each
harness's session files into an estimated cost — an optional invoice line, so
you bill or absorb the AI surplus per client.

### Proof, not a number you typed

A manual entry is a claim a client takes on faith. dira anchors each interval
to commit SHAs that exist in the remote, authored by you — a verifiable line,
not a self-reported one.

## Keep JIRA for tickets. Let dira do the time.

dira is not another tab to remember — it lives in the terminal you are already
in. One command per harness sets up its hooks and you never start a timer
again.

```
dira init   # Claude Code; also codex, gemini, cursor, opencode, grok
```

---

See also [/index.md](https://dirahq.sh/index.md),
[/zavet.md](https://dirahq.sh/zavet.md),
[/getting-started.md](https://dirahq.sh/getting-started.md).