time-tracker

A small, self-hosted time-tracking web app. Log hours per client, task type and period, track daily-hours targets and overtime, and produce printable per-client and per-period reports.

It ships as a single Go binary with server-side rendered templ views, a Bulma UI, and an embedded SQLite database. Static assets and templates are compiled into the binary, so a deployment is one file. That same binary runs the web server (time-tracker serve) and the admin commands (register, export-users).

Features

• Tasks: log time entries with a title, client, task type, period, date and hours. Entries are grouped by day with per-day totals.
• Clients, task types and periods: manage each from its own page, with dedicated edit pages to rename them. A client can restrict which task types it allows, and one period can be marked as the default.
• Daily targets and overtime: set target hours per weekday under My account. The Time page shows each day's hours against its target, grouped by month with monthly subtotals.
• Date-range filter: sum hours, target and overtime between two dates. An open-ended range runs up to today.
• Printable reports: standalone, print-friendly pages that open the browser print dialog for save-as-PDF.
  • /time/report produces a day-by-day time report over the selected range.
  • /clients/{id}/report produces a per-client report: a summary table of hours per task type, followed by one detail table per task type, honoring the active filters.
• Bilingual: English and French (/lang/en, /lang/fr).
• Authentication: email and password login backed by opaque, server-side session tokens stored in a cookie. Sessions live in memory, so they are lost on restart and can be revoked instantly.

Install

Choose one of the options below, then follow First run to create a user and log in.

Docker (recommended)

A multi-arch image is published to GHCR on every push to master and on version tags, for linux/amd64, linux/arm64, linux/arm/v7, linux/arm/v6, linux/386, linux/ppc64le, linux/s390x and linux/riscv64.

docker run -d -p 8080:8080 \
  -e JWT_SECRET=change-me \
  -v time-tracker-data:/data \
  ghcr.io/bloomyindev/time-tracker:latest

The container runs time-tracker serve by default. The database is stored in the /data volume (DB_PATH=/data/time-tracker.db). Run admin commands inside the container with docker exec <container> /app/time-tracker <command> (see CLI).

An example compose.yml is included:

docker compose up -d
docker compose exec time-tracker /app/time-tracker register \
  --email you@example.com --password secret

Prebuilt binary

Download a tarball for your OS and architecture from the Releases page. Each archive contains the single time-tracker binary. Prebuilt targets:

OS	Architectures
Linux	amd64, 386, arm64, armv7, ppc64le, riscv64, s390x, loong64
macOS	amd64, arm64
Windows	amd64, 386, arm64
FreeBSD	amd64, arm64
OpenBSD	amd64, arm64
NetBSD	amd64

tar -xzf time-tracker_*_linux_amd64.tar.gz
cd time-tracker_*_linux_amd64
JWT_SECRET=change-me ./time-tracker serve

From source

Building from source requires Go 1.26 or later. See DEVELOPMENT.md for the full setup.

First run

1. Start the server. It listens on http://localhost:8080. In production, always set JWT_SECRET (see Configuration).
2. Create a user. There is no public sign-up:

   ./time-tracker register --email you@example.com --password secret
   # from source: ./bin/time-tracker register ...
   # docker:      docker exec <container> /app/time-tracker register ...

3. Log in at http://localhost:8080/login with those credentials.

Configuration

The app is configured through environment variables:

Variable	Default	Description
DB_PATH	time-tracker.db	Path to the SQLite database file.
JWT_SECRET	dev-secret-change-me	Secret for signing JWTs used by the (currently unused) bearer-token API flow. Set this in production.

The server always listens on port 8080. The database path can also be passed with --db-path. The schema is created and migrated automatically on startup.

CLI

The admin commands live in the same binary as the server:

time-tracker serve                                           # run the web server
time-tracker register --email <email> --password <password>  # create a user
time-tracker export-users                                    # dump users as JSON (no password hashes)

Every command accepts --db-path (or the DB_PATH environment variable).

Contributing

Development setup, build instructions and the project layout are documented in DEVELOPMENT.md.

License

GNU AGPL v3. See LICENSE.