A single-binary daily work tracker. It embeds a Preact frontend and syncs activity from external sources (GitHub, Jira, Confluence) into a local SQLite database.

• Day-by-day activity log — browse any date and see everything you did across all connected sources in one place
• GitHub — pull requests you authored, reviewed, or commented on; PR statuses (draft → open → in review → approved → merged) kept fresh in the background
• Jira — issues assigned to you that were updated on the day
• Confluence — pages you created, edited, or commented on
• Full-text search — a search bar in the top bar lets you search across all activity and tasks with autocomplete; filter by source (GitHub, Jira, Confluence, or tasks only)
• Dashboard — a dedicated view with stat cards, a stacked bar chart of activity over time, per-source breakdowns, and your most active days; supports preset periods (this week / month / year / all time) and a custom date range
• Background sync — a worker fetches fresh activity on a configurable interval and backfills recent history on startup
• Single binary — the Preact frontend is embedded; no separate web server or database process required
• Markdown backup — optionally mirrors every day to a YYYY/MM/DD.md file tree (tasks + activity, with links) that you can commit to a notes repo or open in any editor
• Local-first — all data is stored in a single SQLite file on your machine; no cloud account needed
• Raycast integration — add tasks to today's list from the Raycast launcher without opening the app

Requires Docker and Docker Compose.

git clone https://github.com/aleksmaksimow/daytracker.git
cd daytracker
cp .env.example .env
$EDITOR .env
docker compose up -d

Open http://localhost:8080. The database is stored in ./data/daytracker.db on the host — it persists across container restarts and rebuilds.

To stop:

docker compose down

To rebuild after a code change:

docker compose up -d --build

Requires Go 1.22+ and Node.js 18+.

git clone https://github.com/aleksmaksimow/daytracker.git
cd daytracker
cd web && npm install && cd ..
make build

This produces a ./daytracker binary with the frontend embedded — no separate web server needed.

Move it somewhere on your PATH so you can run it from anywhere:

sudo mv daytracker /usr/local/bin/daytracker

Requires Go 1.22+ and Node.js 18+.

git clone https://github.com/aleksmaksimow/daytracker.git
cd daytracker
cd web && npm install && cd ..

# Terminal 1 — Go API server
make dev-api

# Terminal 2 — Vite dev server with HMR (proxies /api to :8080)
make dev-web

Open http://localhost:5173.

# Copy and fill in your credentials
cp .env.example .env
$EDITOR .env

# Start the server
./daytracker

Open http://localhost:8080.

make check

Reads your .env and pings each configured connector. Useful for confirming credentials before starting the server.

All configuration is via environment variables prefixed with DAYTRACKER_.

Connectors are enabled automatically when their required variables are set. Unconfigured connectors are silently skipped.

Uses the GitHub GraphQL API with a personal access token.

Required variables:

How to create a token:

1. Go to https://github.com/settings/tokens → Generate new token (classic)
2. Select scopes: repo, read:user
3. Copy the token and set it as DAYTRACKER_GITHUB_TOKEN

What it syncs:

• Pull requests you authored, created on the target date
• Pull requests you reviewed or commented on, updated on the target date (your own PRs are excluded from this list)

PR statuses (open, draft, in review, approved, changes requested, merged, closed) are refreshed every DAYTRACKER_STATUS_REFRESH_INTERVAL for PRs within the DAYTRACKER_BACKFILL_DAYS window.

Uses the Jira REST API v3 with HTTP Basic Auth.

Required variables:

How to create an API token:

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
2. Click Create API token, give it a name, copy the value
3. Set it as DAYTRACKER_JIRA_TOKEN

The same token works for both Jira and Confluence — it is scoped to your Atlassian account, not to a specific product.

What it syncs:

• Issues assigned to you that were updated on the target date

Uses the Confluence REST API v1 with HTTP Basic Auth. The same Atlassian API token used for Jira works here.

Required variables:

See the Jira connector section for instructions on creating an API token.

What it syncs:

• Pages you created or edited on the target date (contributor = currentUser())
• Pages you commented on — multiple comments on the same page are grouped into one activity item

Set DAYTRACKER_BACKUP_DIR to any directory and daytracker will write a snapshot for each synced day:

<backup-dir>/
  2025/
    05/
      27.md
      28.md

Each file contains your tasks (as a checklist) and your activity grouped by source, with titles linked to their original URLs. Files are overwritten on every sync and also refreshed every 2 minutes so task completions land quickly.

The directory is plain text — you can commit it to a notes repo, open it in Obsidian or any Markdown editor, or feed individual day files directly to an AI assistant to answer questions like "what did I work on last Tuesday?" or "summarise my Jira activity this week".

A search bar is always visible in the top bar. Start typing to get an autocomplete dropdown of up to 15 matching activity items and tasks from across all your data.

• Filter by source — use the dropdown to the left of the input to limit results to a specific connector (GitHub, Jira, Confluence) or to tasks only
• Keyboard navigation — use ↑ / ↓ to move through results, Enter to select, Escape to close
• Selecting a result navigates the Journal to the day the item was recorded

Search is powered by SQLite FTS5 with prefix matching, so partial words work (searching "auth" will match "authentication").

Switch to the Dashboard view using the Journal / Dashboard toggle in the top bar.

The dashboard shows a summary of your productivity across a selected time period:

• Stat cards — tasks completed (with completion rate), PRs authored and reviewed, Jira tickets, Confluence pages, and active days
• Activity over time — a stacked bar chart showing GitHub, Jira, Confluence, and task activity; granularity adjusts automatically (daily for ≤60 days, weekly for ≤365 days, monthly beyond that)
• Breakdown by source — per-kind counts for each connector (e.g. authored merged vs open vs draft PRs, Jira done vs in progress)
• Most active days — your top 5 days by total activity volume

Period selection: use the preset chips (This week / This month / This year / All time) or enter a custom date range and click Apply.

Want to add support for a new service? See docs/connectors.md for a full walkthrough of the connector interface, kind naming conventions, frontend wiring, and test patterns.

The raycast-scripts/ directory contains a Raycast script command that lets you add tasks to today's list directly from the Raycast launcher.

Setup:

1. Open Raycast → Settings → Extensions → Script Commands
2. Click Add Directories and select the raycast-scripts/ folder in this repo
3. Search for "Add Daytracker Todo" in Raycast

Usage:

Invoke the command, type your task title, and hit Enter. A notification confirms the task was added. Daytracker must be running for the command to work.

By default the script posts to http://localhost:8080. If you've changed DAYTRACKER_PORT, export it in your shell environment and the script will pick it up.