allenporter · March 1, 2026 16:29 · Jan 31, 2025 · Jan 8, 2025 · Jan 8, 2025 · Jan 8, 2025
diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -52,7 +52,7 @@ to change the IP and port the proxy listens on.
 ## Configure Model Context Protocol Integration
 
 > [!IMPORTANT] 
-NOTE: This integration is currently in review https://github.com/home-assistant/core/pull/135058
+NOTE: This integration is currently available in 2025.2.0b beta release
 
 [![Open your Home Assistant instance and start setting up a new integration.](https://my.home-assistant.io/badges/config_flow_start.svg)](https://my.home-assistant.io/redirect/config_flow_start/?domain=mcp)
 

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -83,19 +83,19 @@ Open the conversation agent and ask it to fetch a web page:
 
 ## Prompt Engineering
 
-Lets now try to use the tool to make a sensor. We should ask the model to respond more succintly:
+Lets now experiment to use the tool to make a sensor. We should ask the model to respond more succintly, and call it from the developer tools. Here is a yaml example:
 
 ```
 action: conversation.process
 data:
   agent_id: conversation.google_generative_ai
   text: >-
-    Please visit bbc.com and summarize the headlines. Please respond with
+    Please visit bbc.com and summarize the first headline. Please respond with
     succinct output as the output will be used as headline for an eInk display
     with limited space. 
 ```
 
-We could even improve this by giving some few-shot example headlines, however, the model follows instructions well already and produces a nice headline:
+We could even improve this by giving some few-shot example headlines, however, the model follows instructions OK already and produces a headline:
 
 ```
 response:

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -145,5 +145,5 @@ template:
 
 <img width="870" alt="Screenshot of template entity output" src="https://gist.github.com/user-attachments/assets/6f085a73-7c80-4cd7-afc5-17c129477495" />
 
-Now go out and make yourself an [ESPHome E-ink display}(https://community.home-assistant.io/t/use-esphome-with-e-ink-displays-to-blend-in-with-your-home-decor/435428) if you don't have one already!
+Now go out and make yourself an [ESPHome E-ink display](https://community.home-assistant.io/t/use-esphome-with-e-ink-displays-to-blend-in-with-your-home-decor/435428) if you don't have one already!
 
diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -113,7 +113,7 @@ response:
 conversation_id: 01JH2B56RMR2DABQQGAAA9X95D
 ```
 
-## Template Entity with Service Responses
+## Template Entity
 
 Below is an example `template` entity definition based on the Model Context Protocol tool calling to fetch a web page:
 
@@ -145,3 +145,5 @@ template:
 
 <img width="870" alt="Screenshot of template entity output" src="https://gist.github.com/user-attachments/assets/6f085a73-7c80-4cd7-afc5-17c129477495" />
 
+Now go out and make yourself an [ESPHome E-ink display}(https://community.home-assistant.io/t/use-esphome-with-e-ink-displays-to-blend-in-with-your-home-decor/435428) if you don't have one already!
+
diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -31,15 +31,14 @@ graph LR
 
 ## Install MCP Proxy & Fetch MCP Server
 
-This will install [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy) and [MCP Fetch Server](https://github.com/modelcontextprotocol/servers/tree/main/src/fetch).
+Install dependencies [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy) and [MCP Fetch Server](https://github.com/modelcontextprotocol/servers/tree/main/src/fetch).
 
-Install dependencies:
 
 ```bash
 $ uv pip install mcp-proxy mcp-server-fetch
 ```
 
-The next step will start an SSE proxy and spawn a local stdio MCP server that can fetch web pages. The proxy
+Most MCP servers are stdio based (e.g. spawned by Claude Desktop) so we need to start a server to expose them to Home Assistant. We use `mcp-proxy` which runs an SSE server, then spawns the stdio MCP server that can fetch web pages. The proxy
 server spawns the command without any environment variables, so we set our path.
 
 ```bash

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -21,7 +21,7 @@ This guide will get you an LLM powered template enttiy using MCP to fetch extern
 graph LR
     A["Home Assistant LLM Conversation Agent"] <--> |sse| B["mcp-proxy"]
     B <--> |stdio| C["mcp-fetch MCP Server"]
-    B <--> |http/s| D["web pages"]
+    C <--> |http/s| D["web pages"]
 
     style A fill:#ffe6f9,stroke:#333,color:black,stroke-width:2px
     style B fill:#e6e6ff,stroke:#333,color:black,stroke-width:2px

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -15,14 +15,18 @@ This assumes you already know about the following:
 
 ## Overview
 
+This guide will get you an LLM powered template enttiy using MCP to fetch external web pages:
+
 ```mermaid
 graph LR
     A["Home Assistant LLM Conversation Agent"] <--> |sse| B["mcp-proxy"]
     B <--> |stdio| C["mcp-fetch MCP Server"]
+    B <--> |http/s| D["web pages"]
 
     style A fill:#ffe6f9,stroke:#333,color:black,stroke-width:2px
     style B fill:#e6e6ff,stroke:#333,color:black,stroke-width:2px
     style C fill:#e6ffe6,stroke:#333,color:black,stroke-width:2px
+    style D fill:#e6ffe6,stroke:#333,color:black,stroke-width:2px
 ```
 
 ## Install MCP Proxy & Fetch MCP Server

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -13,6 +13,18 @@ This assumes you already know about the following:
 - Voice Assistant with Conversation Agent (OpenAI, Google Gemini, Anthropic, etc)
 - Python virtual environment for running MCP server
 
+## Overview
+
+```mermaid
+graph LR
+    A["Home Assistant LLM Conversation Agent"] <--> |sse| B["mcp-proxy"]
+    B <--> |stdio| C["mcp-fetch MCP Server"]
+
+    style A fill:#ffe6f9,stroke:#333,color:black,stroke-width:2px
+    style B fill:#e6e6ff,stroke:#333,color:black,stroke-width:2px
+    style C fill:#e6ffe6,stroke:#333,color:black,stroke-width:2px
+```
+
 ## Install MCP Proxy & Fetch MCP Server
 
 This will install [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy) and [MCP Fetch Server](https://github.com/modelcontextprotocol/servers/tree/main/src/fetch).

diff --git a/Home Assistant Model Context Protocol.md b/Home Assistant Model Context Protocol.md
@@ -3,7 +3,7 @@
 ## TL;DR
 
 Completing these steps will let you have an LLM Powered Web scraper in Home
-Assistant through the [Model Context Protocol](https://github.com/modelcontextprotocol).
+Assistant through the [Model Context Protocol](https://github.com/modelcontextprotocol) with an example of how you could make a template entity for extracting new headlines for a display.
 
 ## Pre-requisites
 

diff --git a/MCP_README.md → Home Assistant Model Context Protocol.md b/MCP_README.md → Home Assistant Model Context Protocol.md
@@ -1,4 +1,4 @@
-# Model Context Protocol
+# Home Assistant Model Context Protocol integration
 
 ## TL;DR
 

diff --git a/MCP_README.md b/MCP_README.md
@@ -2,8 +2,8 @@
 
 ## TL;DR
 
-Completing these steps will let you have an LLM Powered Web Browser in Home
-Assistant using Puppeteer through the [Model Context Protocol](https://github.com/modelcontextprotocol).
+Completing these steps will let you have an LLM Powered Web scraper in Home
+Assistant through the [Model Context Protocol](https://github.com/modelcontextprotocol).
 
 ## Pre-requisites
 
@@ -36,6 +36,9 @@ to change the IP and port the proxy listens on.
 
 ## Configure Model Context Protocol Integration
 
+> [!IMPORTANT] 
+NOTE: This integration is currently in review https://github.com/home-assistant/core/pull/135058
+
 [![Open your Home Assistant instance and start setting up a new integration.](https://my.home-assistant.io/badges/config_flow_start.svg)](https://my.home-assistant.io/redirect/config_flow_start/?domain=mcp)
 
 Manually [add the integration](https://my.home-assistant.io/redirect/config_flow_start/?domain=mcp)

diff --git a/MCP_README.md b/MCP_README.md
@@ -63,5 +63,67 @@ Open the conversation agent and ask it to fetch a web page:
 <img width="396" alt="Screenshot 2025-01-07 at 10 34 10 PM" src="https://gist.github.com/user-attachments/assets/07562834-a35c-4cef-9cef-70796f549ef6" />
 
 
+## Prompt Engineering
 
+Lets now try to use the tool to make a sensor. We should ask the model to respond more succintly:
+
+```
+action: conversation.process
+data:
+  agent_id: conversation.google_generative_ai
+  text: >-
+    Please visit bbc.com and summarize the headlines. Please respond with
+    succinct output as the output will be used as headline for an eInk display
+    with limited space. 
+```
+
+We could even improve this by giving some few-shot example headlines, however, the model follows instructions well already and produces a nice headline:
+
+```
+response:
+  speech:
+    plain:
+      speech: Trump threatens Greenland and Panama Canal.
+      extra_data: null
+  card: {}
+  language: en
+  response_type: action_done
+  data:
+    targets: []
+    success: []
+    failed: []
+conversation_id: 01JH2B56RMR2DABQQGAAA9X95D
+```
+
+## Template Entity with Service Responses
+
+Below is an example `template` entity definition based on the Model Context Protocol tool calling to fetch a web page:
+
+```yaml
+template:
+- trigger:
+    - platform: time_pattern
+      hours: "/1"
+    # Used for testing
+    - platform: homeassistant
+      event: start
+  action:
+    - action: conversation.process
+      data:
+        agent_id: conversation.google_generative_ai
+        text: >-
+          Please visit bbc.com and summarize the single most important headline.
+          Please respond with succinct output as the output will be used as
+          headline for an eInk display with limited space, so answers must be
+          less than 200 characters.
+        response_variable: headline
+  sensor:
+    - name: Headline
+      attributes:
+        title: "{{ headline.response.speech.plain.speech }}"
+      unique_id: "d3641cdf-aa9f-4169-acae-7f7ba989c492"
+  unique_id: "272f0508-3e27-4179-9aca-06d8333874e7"
+```
+
+<img width="870" alt="Screenshot of template entity output" src="https://gist.github.com/user-attachments/assets/6f085a73-7c80-4cd7-afc5-17c129477495" />
 
diff --git a/MCP_README.md b/MCP_README.md
@@ -24,10 +24,10 @@ $ uv pip install mcp-proxy mcp-server-fetch
 ```
 
 The next step will start an SSE proxy and spawn a local stdio MCP server that can fetch web pages. The proxy
-server spawns the command without any environment variables set so we give the full path to uv.
+server spawns the command without any environment variables, so we set our path.
 
 ```bash
-$ mcp-proxy --sse-port 42783 -- /usr/local/bin/uv run mcp-server-fetch
+$ mcp-proxy --sse-port 42783 --env PATH "${PATH}" -- uv run mcp-server-fetch
 ...
 INFO:     Uvicorn running on http://127.0.0.1:42783 (Press CTRL+C to quit)
 ```

diff --git a/MCP_README.md b/MCP_README.md
@@ -56,8 +56,12 @@ The integration will create a new LLM API called `mcp-fetch` that is available t
 
 ## Try it Out
 
-<img width="1440" alt="Screenshot of conversation with agent 1" src="https://gist.github.com/user-attachments/assets/9ce57169-cfea-4e4e-95c5-b7df69fc1ac6" />
+Open the conversation agent and ask it to fetch a web page:
+
+<img width="389" alt="Screenshot 2025-01-07 at 10 33 25 PM" src="https://gist.github.com/user-attachments/assets/68df3c8e-55e7-459e-ba5a-8990fbd6bc80" />
+
+<img width="396" alt="Screenshot 2025-01-07 at 10 34 10 PM" src="https://gist.github.com/user-attachments/assets/07562834-a35c-4cef-9cef-70796f549ef6" />
+
 
-<img width="1440" alt="Screenshot of conversation with agent 2" src="https://gist.github.com/user-attachments/assets/77fb50d8-0910-44a7-97c9-97472757fda3" />
 
 
diff --git a/MCP_README.md b/MCP_README.md
@@ -9,25 +9,25 @@ Assistant using Puppeteer through the [Model Context Protocol](https://github.co
 
 This assumes you already know about the following:
 
-- Node
-- Python
-- Home Assistant
+- Home Assistant 
 - Voice Assistant with Conversation Agent (OpenAI, Google Gemini, Anthropic, etc)
+- Python virtual environment for running MCP server
 
 ## Install MCP Proxy & Fetch MCP Server
 
 This will install [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy) and [MCP Fetch Server](https://github.com/modelcontextprotocol/servers/tree/main/src/fetch).
 
-Install packages
+Install dependencies:
 
 ```bash
-$ ux pip install mcp-proxy mcp-server-fetch
+$ uv pip install mcp-proxy mcp-server-fetch
 ```
 
-This will start an SSE proxy and spawn a local stdio MCP server that can fetch web pages:
+The next step will start an SSE proxy and spawn a local stdio MCP server that can fetch web pages. The proxy
+server spawns the command without any environment variables set so we give the full path to uv.
 
 ```bash
-$ mcp-proxy --sse-port 42783 --env PATH "${PATH}" -- uv run mcp-server-fetch
+$ mcp-proxy --sse-port 42783 -- /usr/local/bin/uv run mcp-server-fetch
 ...
 INFO:     Uvicorn running on http://127.0.0.1:42783 (Press CTRL+C to quit)
 ```
@@ -42,7 +42,7 @@ Manually [add the integration](https://my.home-assistant.io/redirect/config_flow
 
 Set the SSE Server URL to your MCP proxy server SSE endpoint e.g. `http://127.0.0.1:42783/sse`. Make sure the URL ends with `/sse`.
 
-The integration should create a new LLM API called `mcp-fetch`
+The integration will create a new LLM API called `mcp-fetch` that is available to conversation agents. It does not add any other entities or devices.
 
 ## Configure Conversation Agent
 

diff --git a/MCP_README.md b/MCP_README.md
@@ -12,7 +12,7 @@ This assumes you already know about the following:
 - Node
 - Python
 - Home Assistant
-- Conversation Agent (OpenAI, Google Gemini, Anthropic, etc)
+- Voice Assistant with Conversation Agent (OpenAI, Google Gemini, Anthropic, etc)
 
 ## Install MCP Proxy & Fetch MCP Server
 
@@ -52,6 +52,12 @@ The integration should create a new LLM API called `mcp-fetch`
 
 1. Update the prompt to be something simple for now such as: `You are an agent for Home Asisstant, with access to tools through an external server. This external server enables LLMs to fetch web page contents.`
 
-
+<img width="586" alt="Screenshot of Conversation Agent Configuration" src="https://gist.github.com/user-attachments/assets/c4d16919-3ecd-43c3-ad22-17fce32e0a02" />
 
 ## Try it Out
+
+<img width="1440" alt="Screenshot of conversation with agent 1" src="https://gist.github.com/user-attachments/assets/9ce57169-cfea-4e4e-95c5-b7df69fc1ac6" />
+
+<img width="1440" alt="Screenshot of conversation with agent 2" src="https://gist.github.com/user-attachments/assets/77fb50d8-0910-44a7-97c9-97472757fda3" />
+
+
diff --git a/MCP_README.md b/MCP_README.md
@@ -0,0 +1,57 @@
+# Model Context Protocol
+
+## TL;DR
+
+Completing these steps will let you have an LLM Powered Web Browser in Home
+Assistant using Puppeteer through the [Model Context Protocol](https://github.com/modelcontextprotocol).
+
+## Pre-requisites
+
+This assumes you already know about the following:
+
+- Node
+- Python
+- Home Assistant
+- Conversation Agent (OpenAI, Google Gemini, Anthropic, etc)
+
+## Install MCP Proxy & Fetch MCP Server
+
+This will install [mcp-proxy](https://github.com/sparfenyuk/mcp-proxy) and [MCP Fetch Server](https://github.com/modelcontextprotocol/servers/tree/main/src/fetch).
+
+Install packages
+
+```bash
+$ ux pip install mcp-proxy mcp-server-fetch
+```
+
+This will start an SSE proxy and spawn a local stdio MCP server that can fetch web pages:
+
+```bash
+$ mcp-proxy --sse-port 42783 --env PATH "${PATH}" -- uv run mcp-server-fetch
+...
+INFO:     Uvicorn running on http://127.0.0.1:42783 (Press CTRL+C to quit)
+```
+The SSE server is now exposed at `http://127.0.0.1:42783/sse`. You can set flags
+to change the IP and port the proxy listens on.
+
+## Configure Model Context Protocol Integration
+
+[![Open your Home Assistant instance and start setting up a new integration.](https://my.home-assistant.io/badges/config_flow_start.svg)](https://my.home-assistant.io/redirect/config_flow_start/?domain=mcp)
+
+Manually [add the integration](https://my.home-assistant.io/redirect/config_flow_start/?domain=mcp)
+
+Set the SSE Server URL to your MCP proxy server SSE endpoint e.g. `http://127.0.0.1:42783/sse`. Make sure the URL ends with `/sse`.
+
+The integration should create a new LLM API called `mcp-fetch`
+
+## Configure Conversation Agent
+
+1. Navigate to your existing conversation agent integration and reconfigure it
+
+1. Set the *LLM Control* to `mcp-fetch`
+
+1. Update the prompt to be something simple for now such as: `You are an agent for Home Asisstant, with access to tools through an external server. This external server enables LLMs to fetch web page contents.`
+
+
+
+## Try it Out
No results found