Skip to content

kingpanther13/ha-mcp-fork

Β 
Β 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

1,676 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

This is a personal fork. See FORK-DEV.md for the addon-repo workflow.

Breaking change (v7.3.0): ha_config_set_yaml has been moved to beta.

Home Assistant MCP Server Logo

The Unofficial and Awesome Home Assistant MCP Server

95+ Tools Release E2E Tests License
Activity Built with FastMCP Python Version GitHub Sponsors Website

A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with Home Assistant.
Using natural language, control smart home devices, query states, execute services and manage your automations.


Demo with Claude Desktop


πŸš€ Get Started

Full guide to get you started with Claude Desktop (~10 min)

No paid subscription required. Click on your operating system:

Setup Guide for macOSΒ Β Β Β Setup Guide for LinuxΒ Β Β Β Setup Guide for Windows

Quick install (~5 min)

🍎 macOS
  1. Go to claude.ai and sign in (or create a free account)
  2. Open Terminal and run:
    curl -LsSf https://raw.githubusercontent.com/homeassistant-ai/ha-mcp/master/scripts/install-macos.sh | sh
  3. Download Claude Desktop (or restart: Claude menu β†’ Quit)
  4. Ask Claude: "Can you see my Home Assistant?"

You're now connected to the demo environment! Connect your own Home Assistant β†’

🐧 Linux

Anthropic doesn't ship Claude Desktop for Linux, so pick one path:

Claude Desktop β€” free, via the community build:

  1. Install the community Claude Desktop for Linux build and sign in with a free claude.ai account
  2. Open Terminal and run:
    curl -LsSf https://raw.githubusercontent.com/homeassistant-ai/ha-mcp/master/scripts/install-linux.sh | sh
  3. Restart Claude Desktop, then ask: "Can you see my Home Assistant?"

Claude Code β€” official CLI, requires a paid Claude plan:

  1. Install Claude Code: curl -fsSL https://claude.ai/install.sh | bash
  2. Configure ha-mcp, then run claude:
    curl -LsSf https://raw.githubusercontent.com/homeassistant-ai/ha-mcp/master/scripts/install.sh | sh -s -- --claude-code
  3. Start claude, run /mcp to confirm, then ask: "Can you see my Home Assistant?"

Full Linux guide β†’

πŸͺŸ Windows
  1. Go to claude.ai and sign in (or create a free account)
  2. Open Windows PowerShell (from Start menu) and run:
    irm https://raw.githubusercontent.com/homeassistant-ai/ha-mcp/master/scripts/install-windows.ps1 | iex
  3. Download Claude Desktop (or restart: File β†’ Exit)
  4. Ask Claude: "Can you see my Home Assistant?"

You're now connected to the demo environment! Connect your own Home Assistant β†’

🏠 Home Assistant OS (Add-on)
  1. Add the repository to your Home Assistant instance:

    Add Repository

  2. Install "Home Assistant MCP Server" from the Add-on Store and wait for it to complete

  3. Click Start, then open the Logs tab to find your unique MCP URL

  4. Configure your AI client with that URL

No token or credential setup needed β€” the add-on connects to Home Assistant automatically.

Full add-on documentation β†’

🌐 Remote Access (Nabu Casa / Webhook Proxy)

Already have Nabu Casa or another reverse proxy pointing at your Home Assistant? The Webhook Proxy add-on routes MCP traffic through your existing setup β€” no separate tunnel or port forwarding needed.

  1. Install the MCP Server add-on (see above) and the Webhook Proxy add-on from the same store
  2. Start the webhook proxy and restart Home Assistant when prompted
  3. Copy the webhook URL from the add-on logs:
    MCP Server URL (remote): https://xxxxx.ui.nabu.casa/api/webhook/mcp_xxxxxxxx
    
  4. Configure your AI client with that URL

For other remote access methods (Cloudflare Tunnel, custom reverse proxy), see the Setup Wizard.

Webhook proxy documentation β†’

πŸ§™ Setup Wizard for 15+ clients

Claude Code, Gemini CLI, ChatGPT, Open WebUI, VSCode, Cursor, and more.

Open Setup Wizard

Having issues? Check the FAQ & Troubleshooting


πŸ’¬ What Can You Do With It?

Just talk to Claude naturally. Here are some real examples:

You Say What Happens
"Create an automation that turns on the porch light at sunset" Creates the automation with proper triggers and actions
"Add a weather card to my dashboard" Updates your Lovelace dashboard with the new card
"The motion sensor automation isn't working, debug it" Analyzes execution traces, identifies the issue, suggests fixes
"Make my morning routine automation also turn on the coffee maker" Reads the existing automation, adds the new action, updates it
"Create a script that sets movie mode: dim lights, close blinds, turn on TV" Creates a reusable script with the sequence of actions

Spend less time configuring, more time enjoying your smart home.


✨ Features

Category Capabilities
πŸ” Search Fuzzy entity search, deep config search, system overview
🏠 Control Any service, bulk device control, real-time states
πŸ”§ Manage Automations, scripts, helpers, dashboards, areas, zones, groups, calendars, blueprints
πŸ“Š Monitor History, statistics, camera snapshots, automation traces, ZHA devices
πŸ’Ύ System Backup/restore, updates, add-ons, device registry
πŸ”’ Safety Read Only Mode toggle, per-tool enable/disable, tool security policies (user approval), automatic edit backups
Complete Tool List (84 tools)
Category Tools
Add-ons ha_get_addon, ha_manage_addon
Areas & Floors ha_list_floors_areas, ha_remove_area_or_floor, ha_set_area_or_floor
Assist ha_manage_pipeline
Automations ha_config_get_automation, ha_config_remove_automation, ha_config_set_automation
Blueprints ha_get_blueprint, ha_import_blueprint
Calendar ha_config_get_calendar_events, ha_config_remove_calendar_event, ha_config_set_calendar_event
Camera ha_get_camera_image
Dashboard ha_get_dashboard_screenshot (beta)
Dashboards ha_config_delete_dashboard_resource, ha_config_delete_dashboard, ha_config_get_dashboard, ha_config_list_dashboard_resources, ha_config_set_dashboard_resource, ha_config_set_dashboard
Device Registry ha_get_device, ha_remove_device, ha_set_device
Energy ha_manage_energy_prefs
Entity Registry ha_get_entity_exposure, ha_get_entity, ha_remove_entity, ha_set_entity
Files ha_delete_file (beta), ha_list_files (beta), ha_read_file (beta), ha_write_file (beta)
Groups ha_config_list_groups, ha_config_remove_group, ha_config_set_group
HACS ha_get_hacs_info, ha_manage_hacs
Helper Entities ha_config_list_helpers, ha_config_set_helper, ha_remove_helpers_integrations
History & Statistics ha_get_automation_traces, ha_get_history, ha_get_logs
Integrations ha_get_integration, ha_get_system_health, ha_set_integration_enabled
Labels & Categories ha_config_get_category, ha_config_get_label, ha_config_remove_category, ha_config_remove_label, ha_config_set_category, ha_config_set_label
Scenes ha_config_get_scene, ha_config_remove_scene, ha_config_set_scene
Scripts ha_config_get_script, ha_config_remove_script, ha_config_set_script
Search & Discovery ha_get_overview, ha_get_state, ha_search
Service & Device Control ha_bulk_control, ha_call_event, ha_call_service, ha_get_operation_status, ha_list_services
System ha_config_set_yaml (beta), ha_get_updates, ha_manage_backup, ha_manage_custom_tool (beta), ha_manage_theme, ha_reload_core, ha_restart
Todo Lists ha_get_todo, ha_remove_todo_item, ha_set_todo_item
Utilities ha_eval_template, ha_install_mcp_tools (beta), ha_report_issue
Zones ha_get_zone, ha_remove_zone, ha_set_zone

πŸ†š ha-mcp vs. Home Assistant's built-in MCP Server

Home Assistant ships its own MCP Server integration. It is built on the Assist pipeline, so a connected MCP client can read and control the entities you have exposed to Assist and run the intents Assist understands β€” handy for voice-style control of already-exposed devices.

ha-mcp is a standalone server built for configuring, building, and debugging your smart home, not just controlling it. On top of device control, it adds capabilities the built-in integration does not have:

Capability Built-in MCP Server ha-mcp
Control exposed devices, query states Yes Yes
Entity scope Only entities exposed to Assist Everything in Home Assistant
Create / edit automations, scripts, scenes No Yes
Build & edit dashboards No Yes
Debug automations from traces, read history & logs No Yes
Manage helpers, areas, zones, labels, groups No Yes
Backups, add-ons, HACS, device & entity registry No Yes

Rule of thumb: Use the built-in integration for voice-style control of devices you have already exposed; use ha-mcp when you want an AI assistant that can also build and maintain your Home Assistant setup.


πŸ”Œ Custom Component (ha_mcp_tools) (beta)

Some tools require a companion custom component installed in Home Assistant. Standard HA APIs do not expose file system access or YAML config editing. This component provides both.

Tools that require the component:

Tool Description
ha_config_set_yaml (beta) Safely add, replace, or remove top-level YAML keys in configuration.yaml and package files (automatic backup, validation, and config check)
ha_list_files (beta) List files in allowed directories
ha_read_file (beta) Read files from allowed paths (config YAML, logs, and allowed directories)
ha_write_file (beta) Write files to allowed directories
ha_delete_file (beta) Delete files from allowed directories

All other tools work without the component. These five return an error with installation instructions if the component is missing.

These tools also require feature flags: HAMCP_ENABLE_FILESYSTEM_TOOLS=true (file tools) and ENABLE_YAML_CONFIG_EDITING=true (YAML editing). To enable the ha_install_mcp_tools installer tool, set HAMCP_ENABLE_CUSTOM_COMPONENT_INTEGRATION=true.

Install using HACS (recommended)

Open your Home Assistant instance and open a repository inside the Home Assistant Community Store.

To add manually: open HACS > Integrations > three-dot menu > Custom repositories > add https://github.com/homeassistant-ai/ha-mcp (category: Integration) > Download.

After installing, restart Home Assistant. Then open Settings > Devices & Services > Add Integration and search for Home Assistant MCP Server Custom Component.

On Home Assistant OS / Supervised, the integration offers to add the add-on repository and install and start the Home Assistant MCP Server add-on for you β€” no need to add the add-on repository by hand. On Container / Core installs (no Supervisor) there is no add-on; run the server via Docker or pip and the integration just sets up the file/YAML services.

Install manually

Copy custom_components/ha_mcp_tools/ from this repository into your HA config/custom_components/ directory. Restart Home Assistant, then add the integration as described above.


🧠 Better Results with Agent Skills

This server gives your AI agent tools to control Home Assistant. For better configurations, pair it with Home Assistant Agent Skills β€” domain knowledge that teaches the agent Home Assistant best practices.

An MCP server can create automations, helpers, and dashboards, but it has no opinion on how to structure them. Without domain knowledge, agents tend to over-rely on templates, pick the wrong helper type, or produce automations that are hard to maintain. The skills fill that gap: native constructs over Jinja2 workarounds, correct helper selection, safe refactoring workflows, and proper use of automation modes.

Bundled Skills (built-in)

Skills from homeassistant-ai/skills are bundled and served as MCP resources via skill:// URIs. Any MCP client that supports resources can discover them automatically β€” no manual installation needed. For tool-only clients (claude.ai, etc.), the same skills are reachable through the polymorphic ha_get_skill_guide tool β€” call it with no args to list bundled skills, with a skill arg to list its files, or with skill + file to read content. Resources are not auto-injected into context β€” clients must explicitly request them, so idle context cost is just the metadata listing.

ha_get_skill_guide is a mandatory tool: the catalog always exposes it (it can't be disabled) so tool-only clients never see a silently missing skill surface.

Skills can still be installed manually for clients that prefer local skill files β€” see the skills repo for instructions.


πŸ” Tool Discovery for AI Agents

By default, the full tool catalog (~86 tools) is listed to the client through the standard MCP tools/list response. Clients with deferred / on-demand tool loading (Claude Sonnet, Claude Opus,) handle that fine β€” tools are pulled into context only when needed, so idle context cost is near-zero.

For models without deferred tool support β€” Claude Haiku, Gemini, ChatGPT OpenAI-compatible local models, smaller open-weights models β€” listing 86 tools eats ~46K tokens of idle context. To address that, the server ships with a search-based discovery mode built on top of FastMCP's BM25 search transform.

Enable search-based discovery

Set ENABLE_TOOL_SEARCH=true (or toggle the option in the HA add-on). The full catalog is replaced in the tool list with four entry points plus a small set of always-visible "pinned" tools (ha_search_entities, ha_get_overview, ha_restart, etc.). All tools remain callable directly by name once discovered:

Tool Purpose
ha_search_tools BM25 keyword search across all tools. Returns name, description, parameters, and annotations (readOnlyHint / destructiveHint) so the agent can pick the right one.
ha_call_read_tool Execute a readOnlyHint tool by name. Safe β€” clients can auto-approve.
ha_call_write_tool Execute a write tool that creates or updates data.
ha_call_delete_tool Execute a tool that removes / deletes data.

The proxy split lets MCP clients apply different permission policies per category (e.g. auto-approve reads, prompt for writes, confirm deletes) without parsing tool docstrings.

Setting Default Description
ENABLE_TOOL_SEARCH false Replace full tool catalog with search-based discovery (~46K β†’ ~5K idle tokens).
TOOL_SEARCH_MAX_RESULTS 5 Max results returned by ha_search_tools (range 2–10).
PINNED_TOOLS empty Comma-separated tool names to keep always visible. The web settings UI is the primary way to manage this.

When to enable

  • Claude Haiku, OpenAI-compatible local models, Gemini, ChatGPT or any model without native deferred tool support β€” large idle-context savings.
  • MCP clients that cap total tool count (some cap at 100) β€” surfaces a minimal set (~10 tools) instead of 86.
  • Cost-sensitive deployments β€” fewer idle tokens per turn.

Leave it off when using Claude Sonnet/Opus or any client with deferred tool loading; the full catalog has no idle cost there and direct calls skip the search step. If you choose to use our toolsearch then you should disable the native Claude Opus/Sonnet toolsearch, which is called deferred tools in the settings.

πŸ”„ Refresh your client's tool list after changing this (or any) setting. Toggling ENABLE_TOOL_SEARCH (or changing pinned/disabled tools, Read Only Mode, etc.) changes the tools the server exposes, but your AI client keeps serving its cached tool list until it re-fetches. Restarting the add-on or Home Assistant does not refresh the client β€” reconnect or refresh the MCP server in your client (e.g. re-add/refresh the connector in ChatGPT, or close and reopen Claude Desktop). If you skip this, tools shown as available will return Unknown tool when called.

For the HA add-on, the same option is documented in homeassistant-addon/DOCS.md along with the in-add-on settings UI for fine-grained tool enable/disable/pin.


πŸ§ͺ Dev Channel

Want early access to new features and fixes? Dev releases (.devN) are published on every push to master.

Dev Channel Documentation β€” Instructions for pip/uvx, Docker, and Home Assistant add-on.


🀝 Contributing

For development setup, testing instructions, and contribution guidelines, see CONTRIBUTING.md.

For comprehensive testing documentation, see tests/README.md.


πŸ”’ Privacy

Ha-mcp runs locally on your machine. Your smart home data stays on your network.

  • No telemetry today β€” anonymous usage stats are a planned future feature (as of June 2026); when it lands it will follow your Home Assistant analytics/telemetry setting (which you can override), announced prominently in the release notes and the web Settings UI at least one month beforehand
  • No personal data collection β€” we never collect entity names, configs, or device data
  • User-controlled bug reports β€” only sent with your explicit approval

For full details, see our Privacy Policy.


πŸ“„ License

This project is licensed under the MIT License - see the LICENSE file for details.


πŸ™ Acknowledgments

  • Home Assistant: Amazing smart home platform (!)
  • FastMCP: Excellent MCP server framework
  • Model Context Protocol: Standardized AI-application communication
  • Claude Code: AI-powered coding assistant
  • PolicyLayer: Argument-path predicate DSL shape (args.domain in [...] with eq/in/regex/contains/exists/...) inspired the per-tool approval rule schema (#966).

πŸ‘₯ Contributors

Maintainers

Contributors

  • @bigeric08 β€” Explicit mcp dependency for protocol version 2025-11-25 support.
  • @airlabno β€” Support for data field in schedule time blocks.
  • @ryphez β€” Codex Desktop UI MCP quick setup guide.
  • @Danm72 β€” Entity registry tools (ha_set_entity, ha_get_entity) for managing entity properties.
  • @Raygooo β€” SOCKS proxy support.
  • @cj-elevate β€” Integration & entity management tools (enable/disable/delete); person/zone/tag config store routing.
  • @maxperron β€” Beta testing.
  • @kingbear2 β€” Windows UV setup guide.
  • @konradwalsh β€” Financial support via GitHub Sponsors. Thank you! β˜•
  • @knowald β€” Area resolution via device registry in ha_get_system_overview for entities assigned through their parent device. Financial support via GitHub Sponsors. Thank you! β˜•
  • @zorrobyte β€” Per-client WebSocket credentials in OAuth mode, fixing WebSocket tool failures.
  • @deanbenson β€” Fixed ha_deep_search timeout on large Home Assistant instances with many automations.
  • @saphid β€” Config entry options flow tools (initial design, #590).
  • @adraguidev β€” Fix menu-based config entry flows for group helpers (#647).
  • @transportrefer β€” Integration options inspection (ha_get_integration schema support, #689).
  • @teh-hippo β€” Fix blueprint import missing save step.
  • @smenzer β€” Documentation fix.
  • @The-Greg-O β€” REST API for config entry deletion.
  • @restriction β€” Responsible disclosure: python_transform sandbox missing call target validation.
  • @lcrostarosa β€” Diagnostic and health monitoring tools concept (#675), inspiring system/error logs, repairs, and ZHA radio metrics integration.
  • @roysha1 β€” Copilot CLI support in the installation wizard; replaced placeholder logo SVGs with real brand icons on the documentation site.
  • @teancom β€” Fix add-on stats endpoint (/addons/{slug}/stats).
  • @TomasDJo β€” Category support for automations, scripts, and scenes.
  • @bzelch β€” python_transform support for automations and scripts.
  • @gcormier β€” Windows installer improvements: removed unused variable and fixed terminal closing after install.
  • @ekobres β€” Feature flags for HAMCP_ENABLE_FILESYSTEM_TOOLS and HAMCP_ENABLE_CUSTOM_COMPONENT_INTEGRATION in the add-on config, with beta tagging in source and docs.
  • @w3z315 β€” Financial support via GitHub Sponsors. Thank you! β˜•
  • @griffinmartin β€” Added OpenCode (by Anomaly) as a selectable AI client in the setup wizard, with both stdio and streamable HTTP support.
  • @hhopke β€” Fixed addon API calls to route through HA Core ingress proxy instead of direct container connections, fixing ha_manage_addon proxy mode on addon installs.
  • @tomwilkie β€” JMESPath middleware exploration (#1147) whose review-time token-measurement data informed the design of #1199 and #1225.
  • @SealKan β€” fields=/attribute_keys= projection on six read-heavy tools (#1225), ha_call_event tool (#1239), dashboards-list helper refactor (#1207), for:-field duration-math detector in the best-practice checker (#1264), persistent DCR OAuth client registrations across restarts (#1265), and issue-triage prompt token-budgeting (#1522).
  • @KarelTestSpecial β€” Cached YAML instance to prevent CPU spikes during bulk edits (#1371).
  • @corgan2222 β€” HA brand assets for custom integration (#1317).
  • @drseanwing β€” Progress emission via FastMCP Context in long-running tools (#1124); tool-discovery / categorized-search docs (#1123).
  • @fnordpig β€” Config subentry support (#1393) and Assist pipeline management tool (#1392).
  • @paul43210 β€” array_patch mode in ha_manage_addon for atomic GET-modify-POST (#1063).
  • @L1AD β€” Filed #966 proposing tool security policies; pointed to PolicyLayer's MCP-security work as prior art that inspired the predicate DSL shape.
  • @nightcityblade β€” Updated stale Home Assistant Advanced Mode references after HA 2026.6 made formerly advanced options available by default (#1533).
  • @emmelutzer β€” Financial support via GitHub Sponsors. Thank you! β˜•

πŸ’¬ Community


⭐ Star History

Star History Chart

About

The Unofficial and Awesome Home Assistant MCP Server

Resources

License

Contributing

Security policy

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages

  • Python 96.0%
  • Astro 2.5%
  • CSS 0.6%
  • Shell 0.4%
  • JavaScript 0.3%
  • Dockerfile 0.1%
  • PowerShell 0.1%