Skip to main content

AI-powered screen time tracker with intelligent activity analysis and MCP server for AI assistants. Install: pip install telos-tracker && telos setup

Project description

Client

Python desktop application with a terminal UI. Distributed via PyPI as telos-tracker.

Install: pip install telos-tracker && telos setup && telos

What It Does

  • Captures screenshots every 30 seconds
  • Analyzes them with Gemini Vision AI to determine what you're working on
  • Stores structured insights locally in SQLite (screenshots are deleted immediately)
  • Shows a terminal dashboard with timeline, heatmaps, focus scores, and AI chat
  • Syncs data to Firestore for daily email reports (cloud mode)
  • Runs as a background service or interactive TUI

Two modes:

  • Cloud mode -- 7-day free trial, then $3/month. No API key needed.
  • BYOK mode -- Free forever. Bring your own Gemini API key.

Project Structure

client/
├── main.py                    # Entry point (dev), CLI commands, onboarding flow
├── mcp_server.py              # MCP server for Claude Desktop / Cursor
├── service.py                 # Windows background service daemon
├── service_macos.py           # macOS LaunchAgent service
├── core/
│   ├── analyzer.py            # Gemini Vision API integration
│   ├── backend_client.py      # Backend API client (cloud mode)
│   ├── capture.py             # Screenshot capture + activity monitoring
│   ├── daily_aggregator.py    # Daily summary generation, productivity scoring
│   ├── dashboard_server.py    # Local web dashboard (Flask, localhost:5555)
│   ├── database.py            # SQLite operations
│   ├── email_reporter.py      # Legacy email reporter (SMTP)
│   ├── fallback_handler.py    # Backend → local Gemini → offline fallback
│   ├── firebase_auth.py       # Firebase authentication
│   ├── firestore_sync.py      # Background sync: SQLite → Firestore
│   ├── goal_manager.py        # Analysis goals management
│   ├── onboarding.py          # First-run state management
│   ├── query_engine.py        # AI chat engine
│   ├── session_builder.py     # Group captures into work sessions
│   └── trial_manager.py       # Trial period tracking
├── tui/
│   ├── app.py                 # Main Textual TUI application
│   ├── screens/               # All TUI screens (dashboard, timeline, chat, etc.)
│   ├── widgets/               # Custom widgets (heatmap, waveform, breakdown)
│   ├── workers/               # Background workers (capture, sync, email)
│   └── models/                # TUI state management
├── utils/
│   ├── config_manager.py      # YAML config loading and saving
│   ├── hash_utils.py          # Perceptual hashing for duplicate detection
│   ├── prompt_loader.py       # AI prompt template loading
│   └── sentry_utils.py        # Sentry error tracking
├── telos_tracker/
│   ├── __init__.py            # Package version
│   └── cli.py                 # CLI entry point (telos command)
├── prompts/                   # AI prompt templates (editable)
│   ├── screenshot_analysis.txt
│   ├── session_enrichment.txt
│   ├── daily_summary.txt
│   └── ai_chat_system.txt
├── pyproject.toml             # PyPI package config
├── requirements.txt           # Python dependencies
├── publish.ps1                # Windows PyPI publish script
├── publish.sh                 # Linux/macOS PyPI publish script
├── build_installer.py         # Windows executable builder (PyInstaller)
├── build_macos.py             # macOS .app bundle builder
└── config.yaml.example        # Configuration template

Local Development

cd client
pip install -r requirements.txt
python main.py setup           # Interactive setup wizard
python main.py                 # Launch TUI

Other commands:

python main.py test            # Test capture loop
python main.py stats           # Show statistics
python main.py service-console # Run as background daemon

TUI Keyboard Shortcuts

Key Screen
D Dashboard
T Timeline
S Summary
A AI Chat
C Settings
E Email settings (from settings)
G Goals
H Help
Q Quit

Configuration

Created by telos setup at ~/.telos/config.yaml. Key settings:

Setting Default Description
capture.interval_seconds 30 Screenshot interval
capture.idle_threshold 300 Seconds before marking idle
backend.enabled true (cloud) Use cloud backend for analysis
backend.url Cloud Run URL Backend API endpoint
email.enabled true Daily email reports
email.send_time 21:00 When to send daily report

Publishing to PyPI

# Automated (handles version bump, build, upload, tag)
.\publish.ps1 0.2.7 "Description of changes"

# Manual
python -m build
twine upload dist/*

Building Desktop Apps

python build_installer.py      # Windows → dist/Telos.exe
python build_macos.py          # macOS → dist/Telos.app

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

telos_tracker-0.2.8.tar.gz (366.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

telos_tracker-0.2.8-py3-none-any.whl (219.2 kB view details)

Uploaded Python 3

File details

Details for the file telos_tracker-0.2.8.tar.gz.

File metadata

  • Download URL: telos_tracker-0.2.8.tar.gz
  • Upload date:
  • Size: 366.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.4

File hashes

Hashes for telos_tracker-0.2.8.tar.gz
Algorithm Hash digest
SHA256 85e885eca5077f75892ccc92f0483c00cd5fd4e13009c3d8ff36593514a1dc89
MD5 598769c188c758fbdb78d3cfa8a77f79
BLAKE2b-256 940801eea97fc832c8ebaf9e0520e8bbca49b2672ccde69f52c889edd470917a

See more details on using hashes here.

File details

Details for the file telos_tracker-0.2.8-py3-none-any.whl.

File metadata

  • Download URL: telos_tracker-0.2.8-py3-none-any.whl
  • Upload date:
  • Size: 219.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.4

File hashes

Hashes for telos_tracker-0.2.8-py3-none-any.whl
Algorithm Hash digest
SHA256 07a485d487bd9efd901dbc90b8774008b8399df327af1b17f8fd9f0499645c0d
MD5 8c25ce9149d77f5332f7ce78f9e8ff32
BLAKE2b-256 c41344b165f3a8e4f7d5e6569f3c1e9e8de6c1a5667cfa2ba00bd9e012208394

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page