Data Points

Record meaningful events in Home Assistant and analyze them directly alongside your entity history, long-term statistics, and chart annotations.

Website

See the v0.6.0 release for screenshots of the anomaly monitor wizard and AI query brief.

---

Overview

Data Points is a Home Assistant integration for recording timestamped events and using them as analytical context across charts, lists, and a dedicated history page.

It helps you answer questions like:

• What changed?
• When did it change?
• What system or entity was affected?
• Was the change expected?
• Does the related sensor behavior now look suspicious?

The integration bundles its Lovelace cards and panel frontend automatically. No separate Lovelace resource configuration is required.

---

Why Data Points is useful

A plain chart tells you what changed. Data Points helps you understand why it changed by combining:

• raw measurements
• long-term statistics
• user-created or automation-created annotations
• target-aware chart overlays
• anomaly detection
• historical date-window comparison

That makes it much easier to investigate heating behavior, energy usage, sensor faults, maintenance effects, occupancy-driven changes, and operational regressions over time.

---

What Data Points provides

• Record custom datapoints from automations, scripts, dashboards, and Developer Tools
• Attach datapoints to entities, devices, areas, or labels
• Render datapoints directly on history, statistics, and sensor charts
• Browse, search, edit, delete, and hide datapoints in a dedicated list card
• Investigate entity history with target rows, per-target analysis options, and date-window comparisons
• Create chart annotations directly from the chart while exploring data
• Backend-powered anomaly detection to highlight suspicious behavior, including rate-of-change (point_to_point) detection across eight configurable window sizes
• Anomaly monitors appear as Home Assistant integration devices, enabling automations to trigger when anomalies are detected
• Compare a current period against saved historical windows to find drift and regressions
• AI query brief: a structured panel context for AI assistants (MCP-connected or otherwise) containing entity metadata, query inputs, and current anomaly findings — localized into DE, ES, FR, PT, FI, and ZH-Hans
• Native MCP tools expose chart analysis, annotations, and monitor findings through Home Assistant 2026.11 or later.

---

Documentation

Topic	Description
Cards & UI	All six cards, the history panel, and YAML configuration examples
Recording Datapoints	The record action, fields, automation patterns, and examples
History & Analysis	Chart features, trend methods, anomaly detection, and date windows
Development	Setup, build, tests, i18n, remote HA dev, WebSocket API, and CI

---

Translations

Data Points ships with both Home Assistant integration translations and frontend card/panel translations.

Included locales

• English
• Finnish
• French 🤖
• German 🤖
• Spanish 🤖
• Portuguese 🤖
• Simplified Chinese 🤖

English is the source language. Finnish translations were written by a non-native speaker. All other bundled locales are machine-translated — they are usable defaults rather than fully reviewed translations. Improvements are very welcome.

• Integration/service strings: custom_components/hass_datapoints/translations/*.json
• Frontend component strings: custom_components/hass_datapoints/src/**/i18n/

---

Roadmap

Planned features

• Multiple saved views — persist more than one named chart-and-panel state for reusable investigation setups.
• Automatic historical period matching — find similar historical periods automatically so the panel can suggest or create date windows.
• Automatic anomaly-to-datapoint generation — generate datapoints when configured anomaly conditions are met. (Anomaly monitors are now HA entities, so automation triggers are already possible — this extends that to automatic datapoint creation.)
• Backfilling tools — create datapoints from recent history and long-term statistics after the fact.
• Anomalies summary card — dedicated card for highlighting current anomalies with deep-links into the history view.
• Drop-in replacements for HA sensor and statistics cards — equivalents that support datapoint overlays and richer contextual controls.

Themes

• Operational memory — preserve and reuse saved investigative contexts.
• Assisted comparison — reduce the manual work needed to find meaningful historical baselines.
• From analysis to action — let anomaly configuration graduate into automations and auto-generated datapoints.
• Dashboard-native investigation — bring anomaly surfacing into smaller cards for everyday dashboards.

---

Installation

HACS

1. Open HACS.
2. Go to Integrations.
3. Add this repository as a custom repository with category Integration.
4. Install Data Points.
5. Restart Home Assistant.

Manual

Copy custom_components/hass_datapoints into:

config/custom_components/hass_datapoints

Then restart Home Assistant.

---

Setup

Add the integration from:

Settings → Devices & Services → Add Integration → Data Points

No YAML setup is required.

---

Quick start

Record a datapoint from Developer Tools → Actions:

action: hass_datapoints.record
data:
  message: "Heating schedule changed"
  entity_ids:
    - climate.living_room
    - sensor.living_room_temperature
  icon: mdi:radiator
  color: "#ff5722"

Then add the history card to a dashboard:

type: custom:hass-datapoints-history-card
title: Living room
entities:
  - sensor.living_room_temperature
hours_to_show: 72

See Recording Datapoints and Cards & UI for the full reference.