Mtp Hot Reload logo

Mtp Hot Reload

OrganizationPopular
dotnet
mtp-hot-reload

Set up or recover MTP hot reload for a long-lived console-host edit/re-run loop in a Microsoft Testing Platform project. Use for explicit MTP console requests such as "enable hot reload", "dotnet run to rerun on edit", or unsupported/rude edits. Covers setup, run/watch, restarts, filters, and the VSTest no-mutation fallback. For one-time runs, exact commands, filter errors, TRX/dumps, or merely a failing test, use run-tests. Do not use when Test Explorer or an IDE should rerun tests without an MTP console host. Excludes editor integration, CI, and writing/debugging tests.

Overview

Publisherdotnet
Repositoryskills
Skill namemtp-hot-reload
Stars
5.5K
Forks
420
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by dotnet on GitHub. Read the source before you install it.

Installation

Install the Mtp Hot Reload AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/dotnet/skills/tree/main/plugins/dotnet-test/skills/mtp-hot-reload
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/dotnet/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/plugins/dotnet-test/skills/mtp-hot-reload .claude/skills/mtp-hot-reload
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Mtp Hot Reload in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Mtp Hot Reload on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Mtp Hot Reload is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

MTP Hot Reload for Iterative Test Fixing

Set up and use a long-lived Microsoft Testing Platform host that applies code edits and automatically reruns tests.

When to Use

  • User explicitly asks for test hot reload
  • User wants a host to stay running and automatically rerun after repeated edits
  • User needs to set up MTP hot reload in their project

When Not to Use

  • User needs to write new tests from scratch (use general coding assistance)
  • User needs to diagnose why a test is failing (use diagnostic skills)
  • User asks about Visual Studio or VS Code Test Explorer hot reload or an IDE-integrated rerun experience (different feature, not MTP console-host hot reload)
  • User asks whether the editor can auto-rerun affected tests after source edits without using an MTP console host
  • User wants one normal run without rebuilding (use run-tests)
  • User needs CI/CD pipeline configuration

Inputs

InputRequiredDescription
Test project pathNoPath to the test project (.csproj). Defaults to current directory.
Failing test name or filterNoSpecific test(s) to iterate on

Response sizing

  • If setup is already complete and the user asks only which command to use, return one dotnet run --project <path> command and one sentence explaining that it starts the persistent host. Do not repeat package, launch profile, or rude-edit guidance.
  • If the package is already installed, show only the remaining enable-and-run steps. Do not suggest reinstalling it or add optional persistence/recovery paths unless requested.
  • For a named test, identify the framework and return one runnable command with that framework's filter syntax. Never substitute MSTest/NUnit --filter for xUnit v3 --filter-method or TUnit --treenode-filter.
  • A directly launched MTP host is itself the persistent watcher: dotnet run stays alive and the HotReload extension reacts to edits. Do not replace it with dotnet watch run for the normal MTP path; reserve dotnet watch for the explicit restart fallback below.

Workflow

Step 1: Detect the platform before changing anything

Hot reload requires MTP. It does not work with VSTest.

Follow the complete evaluated-property procedure in the platform-detection skill. Read imported props and package versions as well as the project file. Do this before installing packages, editing files, or returning an MTP launch command.

Hard stop for VSTest: report that MTP hot reload is unavailable for the project as configured and stop the MTP setup path. Do not install the extension, create launchSettings.json, set the environment variable, change runner properties/packages, or return a dotnet run hot-reload command. Never turn a setup request into an implicit VSTest-to-MTP migration.

Offer one valid non-MTP fallback that preserves the project:

shell
dotnet watch --project <project-path> test

This rebuilds and reruns the existing VSTest project when files change; it is not MTP hot reload. Offer an explicit migration as a separate option, but do not perform it unless the user asks. Exact one-shot test commands remain owned by run-tests.

Step 2: Add the hot reload NuGet package

First inspect the effective package references. If Microsoft.Testing.Extensions.HotReload is already installed, preserve its version and skip this step. Otherwise install it:

shell
dotnet add <project-path> package Microsoft.Testing.Extensions.HotReload

Note: When using Microsoft.Testing.Platform.MSBuild (included transitively by MSTest, NUnit, and xUnit runners), the extension is auto-registered when you install its NuGet package -- no code changes needed.

Step 3: Enable hot reload

Hot reload is activated by setting the TESTINGPLATFORM_HOTRELOAD_ENABLED environment variable to 1.

Option A -- Set it in the shell before running tests:

shell
# PowerShell
$env:TESTINGPLATFORM_HOTRELOAD_ENABLED = "1"

# bash/zsh
export TESTINGPLATFORM_HOTRELOAD_ENABLED=1

Option B -- Add it to launchSettings.json (recommended for repeatable use):

Create or update Properties/launchSettings.json in the test project:

json
{
  "profiles": {
    "<ProjectName>": {
      "commandName": "Project",
      "environmentVariables": {
        "TESTINGPLATFORM_HOTRELOAD_ENABLED": "1"
      }
    }
  }
}

Step 4: Run the tests with hot reload

Run the test project directly (not through dotnet test) to use hot reload in console mode:

shell
dotnet run --project <project-path>

To filter to specific failing tests, pass the filter after --. The syntax depends on the test framework -- see the filter-syntax skill for full details. Quick examples:

FrameworkFilter syntax
MSTestdotnet run --project <path> -- --filter "FullyQualifiedName~TestMethodName"
NUnitdotnet run --project <path> -- --filter "FullyQualifiedName~TestMethodName"
xUnit v3dotnet run --project <path> -- --filter-method "*TestMethodName"
TUnitdotnet run --project <path> -- --treenode-filter "/*/*/ClassName/TestMethodName"

The test host will start, run the tests, and remain running waiting for code changes.

Step 5: Iterate on the fix

  1. Edit the source code (test code or production code) in your editor
  2. The test host detects the changes and re-runs the affected tests automatically
  3. Review the updated results in the console
  4. Repeat until all targeted tests pass

Important: Hot reload currently works in console mode only. There is no support for hot reload in Test Explorer for Visual Studio or Visual Studio Code.

Unsupported edits and rude edits

Method-signature changes, new types, and other unsupported edits cannot be applied to the active process. Never imply that the stale host picked them up.

For a directly launched MTP host:

  1. Preserve the exact command, profile, environment, filter, and arguments that started the current host.
  2. Stop it with Ctrl+C.
  3. Rebuild the same project: dotnet build <project-path>.
  4. Rerun the same original host command. Do not replace an unknown existing invocation with a generic dotnet run command.

If repeated unsupported edits are expected, offer a watch-managed restart fallback:

shell
# PowerShell
$env:TESTINGPLATFORM_HOTRELOAD_ENABLED = "1"
$env:DOTNET_WATCH_RESTART_ON_RUDE_EDIT = "1"
dotnet watch --project <project-path> run -- <existing-MTP-arguments>

dotnet watch restarts the process when a rude edit cannot be applied. Without the auto-restart variable, accept the restart prompt or press Ctrl+R. Preserve any existing test filter after --. DOTNET_WATCH_RESTART_ON_RUDE_EDIT is a documented .NET SDK switch, not an MTP extension setting. When correctness may be questioned, cite the dotnet watch documentation and say that it makes the watcher restart instead of prompting. Never omit the original filter or replace the requested watch-managed fallback with a one-shot dotnet run.

Step 6: Finalize

Once all tests pass:

  1. Stop the test host (Ctrl+C)
  2. Use run-tests when the user requests an exact one-shot validation command, flags, filter, TRX, or dump
  3. Optionally remove TESTINGPLATFORM_HOTRELOAD_ENABLED from the environment or keep launchSettings.json for future use

Validation

  • Project uses Microsoft Testing Platform (not VSTest)
  • Microsoft.Testing.Extensions.HotReload package is installed
  • TESTINGPLATFORM_HOTRELOAD_ENABLED environment variable is set to 1
  • Tests run and the host remains active waiting for changes
  • Code changes are picked up without manual restart

Common Pitfalls

PitfallSolution
Using dotnet test instead of dotnet runHot reload requires dotnet run --project <path> to run the test host directly in console mode
Project uses VSTest, not MTPDo not mutate it. Offer dotnet watch --project <path> test as a rebuild/rerun fallback or a separate explicit migration
Forgetting to set the environment variableSet TESTINGPLATFORM_HOTRELOAD_ENABLED=1 before running
Expecting Test Explorer integrationConsole mode only -- no VS/VS Code Test Explorer support
Making unsupported code changes (rude edits)Stop, rebuild, and rerun the same host invocation, or use dotnet watch with restart-on-rude-edit behavior

Frequently asked questions

What does the Mtp Hot Reload AI skill do?

Set up or recover MTP hot reload for a long-lived console-host edit/re-run loop in a Microsoft Testing Platform project. Use for explicit MTP console requests such as "enable hot reload", "dotnet run to rerun on edit", or unsupported/rude edits. Covers setup, run/watch, restarts, filters, and the VSTest no-mutation fallback. For one-time runs, exact commands, filter errors, TRX/dumps, or merely a failing test, use run-tests. Do not use when Test Explorer or an IDE should rerun tests without an MTP console host. Excludes editor integration, CI, and writing/debugging tests.

Why use Mtp Hot Reload on TypingMind?

Because you install it once and use it with any model. Mtp Hot Reload is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Mtp Hot Reload in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/dotnet/skills/tree/main/plugins/dotnet-test/skills/mtp-hot-reload. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Mtp Hot Reload?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Mtp Hot Reload?

As many as you like. As long as a model supports skills, you can use Mtp Hot Reload with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Mtp Hot Reload AI skill free?

Yes. It is published on GitHub by dotnet under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇