Privacy & Data
What data do you collect?
Only what's needed to run the service and improve source reliability. Here's the complete list — nothing else is collected.
No account data is collected. All features work without registration.
Capability telemetry (anonymous, GDPR-compliant — see details below)
| Field | Type | Example | Purpose |
|---|---|---|---|
| install_uuid | string | a1b2c3d4-... | Count unique installs (auto-generated, not tied to identity) |
| product | string | "energy" | Know which product is installed |
| country | string | "NL" | Know which bidding zones are active |
| ha_version | string | "2025.2.1" | Compatibility tracking |
| addon_version | string | "1.2.0" | Know which versions are in the field |
| entity_count | integer | 847 | Understand typical install sizes |
| energy_provider | string | "tibber" | Supplier coverage insights |
| os_arch | string | "aarch64" | Determine build target priority |
| ha_install_type | string | "os" | Know which HA variants to support |
| supervisor_channel | string | "stable" | Compatibility risk assessment |
| locale | string | "nl" | Localization priority (language, not location) |
| active_source | string | "worker" | Monitor which price source tier is active (worker, energy_charts, cache) |
| uptime_bucket | bucket | ">30d" | Addon stability monitoring |
| fallback_bucket | bucket | "0" | Detect price source reliability issues |
| cache_hit_bucket | bucket | ">90%" | Measure offline resilience |
| cleanup_deleted_bucket | bucket | "25-100" | Measure cleanup effectiveness (Care only) |
| cleanup_duration_bucket | bucket | "<1s" | Monitor cleanup performance per hardware class (Care only) |
Buckets, not exact values. Fields like uptime, fallback count, and cache hit ratio are sent as ranges (e.g., ">30d", "0", ">90%"), never as exact numbers. This prevents usage fingerprinting while still providing product insights.
90-day retention. Telemetry from installations inactive for 90+ days is automatically deleted.
No passwords stored. No browser fingerprinting. No analytics scripts. No third-party trackers. No usage patterns. No timestamps of individual actions. No data sold or shared. Ever.
Does any Home Assistant data leave my network?
No. All diagnostics, health scans, database analysis, and cleanup operations run entirely on your local machine. The addon communicates with your HA instance over its local Supervisor API. The only outbound connections are:
• Price data fetches — Energy app fetches prices from the Synctacles API (Energy-Charts + Elexon + ENTSO-E fallback)
• KB search — Care app searches the knowledge base via
• Telemetry — anonymous, for product improvement only
• AI diagnostics — Care AI goes directly to your chosen provider's API (Anthropic, OpenAI, Groq or Gemini) using your own key (we never see these conversations)
• Price data fetches — Energy app fetches prices from the Synctacles API (Energy-Charts + Elexon + ENTSO-E fallback)
• KB search — Care app searches the knowledge base via
api.synctacles.com (Cloudflare edge)• Telemetry — anonymous, for product improvement only
• AI diagnostics — Care AI goes directly to your chosen provider's API (Anthropic, OpenAI, Groq or Gemini) using your own key (we never see these conversations)
Is telemetry enabled by default?
Yes. Telemetry is always enabled and GDPR-compliant. It sends anonymous capability data — system metadata (architecture, HA version, locale) and operational health (active price source, cache efficiency, uptime range). All values that could fingerprint usage are sent as buckets (ranges), not exact numbers. See the complete field list above.
It contains no personally identifiable information. No IP addresses, no usage timestamps, no automation names, no energy consumption data. We collect system state, not user behavior.
It contains no personally identifiable information. No IP addresses, no usage timestamps, no automation names, no energy consumption data. We collect system state, not user behavior.
Do you store my IP address?
Not in plain text. We hash IP addresses with SHA-256 before storage. The hash is used solely for rate limiting (preventing abuse). The original IP cannot be recovered from the hash.
Transparency note: Under GDPR, hashed IPs are considered pseudonymized data, not fully anonymous data. We treat them accordingly: hashes are stored only for the duration needed for rate limiting, are never combined with other data to identify individuals, and are automatically purged. We do not store user-agent strings or any other network metadata.
Transparency note: Under GDPR, hashed IPs are considered pseudonymized data, not fully anonymous data. We treat them accordingly: hashes are stored only for the duration needed for rate limiting, are never combined with other data to identify individuals, and are automatically purged. We do not store user-agent strings or any other network metadata.
Architecture & Privacy
What connections does Synctacles make?
Energy fetches real-time prices from the Synctacles API and caches them locally for 48 hours. Care runs all health scans and security audits on your system; KB search goes via
Your Home Assistant entities, automations, logs, and configuration never leave your network.
api.synctacles.com (Cloudflare edge). Both apps send anonymous telemetry for product improvement.Your Home Assistant entities, automations, logs, and configuration never leave your network.
Where does price data come from?
Price data covering 50 zones across 31 countries, from 5 independent data sources:
Energy-Charts (EU wholesale prices, CC BY 4.0), Elexon BMRS (GB wholesale prices), ENTSO-E (fallback), Elering (Baltics/Finland), and zone correlation (estimation for remaining zones).
The app delivers prices through a 3-tier delivery chain: Synctacles API → Energy-Charts (app-side direct) → local 48h cache. The “5-tier fallback” on the homepage refers to the 5 data sources; the “3-tier” here refers to how the app fetches and caches those prices.
The app delivers prices through a 3-tier delivery chain: Synctacles API → Energy-Charts (app-side direct) → local 48h cache. The “5-tier fallback” on the homepage refers to the 5 data sources; the “3-tier” here refers to how the app fetches and caches those prices.
What happens if a price source goes down?
The app has a 3-tier fallback system: Synctacles API (Energy-Charts primary, ENTSO-E fallback) → Energy-Charts (app-side direct) → local 48-hour cache. If the primary source fails, it automatically falls through the tiers. Your automations keep running on cached data.
How do addons communicate with Home Assistant?
Via MQTT Discovery for sensors (auto-discovered, zero config) and the HA Supervisor API for system info (version, backups, add-on status). Both are local protocols — no external network traffic.
Security
Is the code open source?
All active repositories are on GitHub at github.com/synctacles. The compiled Go binaries are distributed via the HA app store.
Does Care modify my HA database automatically?
Never without explicit consent. Care runs read-only diagnostics by default. Cleanup actions (orphan removal, vacuum) require you to click a button and confirm. Before any write operation, Care checks that a recent HA backup exists.
How are updates delivered?
Through the standard Home Assistant add-on update mechanism. When we push a new version to our GitHub repository, HA detects it and offers the update in your add-on store. You control when to update. No auto-updates, no forced upgrades.
Licensing & Cost
Is it really free?
Yes. All features. Both products. No trial. No feature gates. No "upgrade to unlock" prompts.
Will there be a paid tier?
We currently have no plans for paid tiers. If we ever introduce additional features, our intention is not to paywall what is currently free.
How is this sustainable?
Low infrastructure cost. The apps run on your hardware. Our backend is a Cloudflare Worker that handles KB search and telemetry — minimal cost at any scale. No heavy compute, no ML inference, no media storage.
Longevity
What happens if Synctacles shuts down?
Your apps keep working. Energy falls back to cached data and direct source fetching. Care diagnostics run on your system. KB search and price data require internet, but both degrade gracefully with local caching.
Is this dependent on a single server?
No. The backend runs as a Cloudflare Worker — globally distributed with automatic failover. But the apps are designed to function independently. Energy has a 48-hour local cache and can fetch prices directly from public sources. Care diagnostics run entirely offline. The backend is a convenience, not a dependency.
Can I fork it?
The source code is on GitHub. Standard open-source rules apply.