Runiq.AI.Cli 1.0.0

Runiq CLI

Runiq CLI creates ready-to-run ASP.NET Core projects for Runiq AI. It sets up a Visual Studio friendly solution, adds Runiq packages through NuGet, wires the generated API project, and can include a small starter sample with agents, typed tools, a real Runiq workflow, Dashboard, and MCP.

Installation

Install the CLI as a .NET tool:

dotnet tool install --global Runiq.AI.Cli --version 1.0.0

Update an existing installation:

dotnet tool update --global Runiq.AI.Cli --version 1.0.0

Verify the command is available:

runiq --help
runiq --version

Create A Project

Run:

runiq init

Or provide the project name up front:

runiq init MyRuniqApp

The wizard asks for:

• Project name, when it is not provided in the command
• Default AI provider
• Provider API key setup
• Starter sample code
• Dashboard
• MCP

Supported provider choices:

• OpenAI
• Azure OpenAI
• Ollama
• Anthropic

You can skip API key setup during generation and configure it later. If you choose to enter a key, the CLI stores it with .NET user secrets for the generated API project.

Generated Output

For a project named Sample04, the CLI creates a solution like this:

Sample04/
  Sample04.sln
  README.md
  src/
    Sample04.Api/
      Sample04.Api.csproj
      Program.cs
      Agents/
      Tools/
      Flows/
      Mcp/
  tests/

The starter artifacts live inside src/{ProjectName}.Api, so they are visible directly under the API project in Visual Studio.

Generated projects use NuGet package references only. They do not use local Runiq source or project references.

Run The Generated API

After generation:

cd MyRuniqApp
dotnet run

If Dashboard is enabled, open:

https://localhost:{port}/dashboard

If MCP is enabled, the MCP endpoint is available at:

https://localhost:{port}/mcp

The CLI prints the detected Dashboard and MCP URLs when the generated launch settings expose an application URL.

Starter Sample

When starter sample code is enabled, the generated project includes a compact version of the repository's Runiq.AI.WorkflowTravelPlanner scenario.

Try this in the Dashboard:

Create a practical one-day historical trip plan in Istanbul for two people. Keep it easy to walk.

The sample includes:

• WeatherAgent
• PlacesAgent
• PlannerAgent
• WeatherTool
• PlacesTool
• MealSuggestionTool
• TravelPlanningFlow.cs
• Mcp/README.md when MCP is enabled

The conceptual flow is:

WeatherAgent + WeatherTool
  -> PlacesAgent + PlacesTool
  -> PlannerAgent + MealSuggestionTool
  -> final itinerary

The sample is intentionally small. It is meant to show where agents, tools, prompts, and workflows belong without adding large datasets or complex business logic.

Sample Tools

WeatherTool returns a simple hardcoded weather response:

Istanbul weather is mild and partly cloudy, around 18 C.

PlacesTool returns a small deterministic set of walkable places, while MealSuggestionTool keeps meal stops close to the route.

Sultanahmet Square -> Gulhane Park -> Karakoy

MCP

When MCP is enabled, the generated project references Runiq.AI.Mcp, registers MCP services, and maps /mcp.

The starter tools include MCP metadata for:

• weather.get
• places.get
• meal.suggest

The generated Mcp/README.md explains the intended MCP tool examples.

Generated Project README

The generated project README is user-facing. It gives prompt examples instead of documenting internal folder structure.

Example prompts:

Create a practical one-day historical trip plan in Istanbul for two people. Keep it easy to walk.

Plan the same trip for Izmir and account for the weather.

Which agents and tools contributed to the result?

Show the workflow trace for the Istanbul plan.

Notes

• The CLI currently provides the init command.
• The wizard is interactive.
• Generated projects are intended to run immediately after creation.
• Starter workflow output is a registered Runiq.AI.Workflows flow, matching the repository sample architecture.
• The starter sample does not create helper, utility, service, CRM, order, or customer examples.

Related Packages

Runiq AI is modular. Generated projects may use:

• Runiq.AI.Core
• Runiq.AI.Workflows
• Runiq.AI.Mcp