Generate the code from typespec for C#. Parameter: C# SDK repository root location <cs_root>.
npx skills add https://github.com/Azure/azure-sdk-for-net --skill generate-code-cs
The C# repository root location is provided by <cs_root>. C# source codes of interest are located in <cs_root>/sdk/ai/Azure.AI.Projects, <cs_root>/sdk/ai/Azure.AI.Projects.Agents and <cs_root>/sdk/ai/Azure.AI.Extensions.OpenAI. Each folder contains subfolder src, which contains all the source codes. The src/Generated folder contains generated code; the folder Custom contains the code customizations. Please review the references/customization.md for more information on code customization. The typespec repository, used for code generation is located at https://github.com/Azure/azure-rest-api-specs.git. Please use the branch provided by user, if it is not provided use feature/foundry-release.
commit field in three files: <cs_root>/sdk/ai/Azure.AI.Projects.Agents/tsp-location.yaml, <cs_root>/sdk/ai/Azure.AI.Projects/tsp-location.yaml and <cs_root>/sdk/ai/Azure.AI.Extensions.OpenAI/tsp-location.yaml.cd <cs_root>/sdk/ai/Azure.AI.Projects.Agents
dotnet build /t:GenerateCode
cd <cs_root>/sdk/ai/Azure.AI.Projects.Agents
dotnet build /t:GenerateCode
If the EPERM error has happened, run the next code and make sure it passes:
cd ../Azure.AI.Extensions.OpenAI
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Projects.Agents\src/../
cd ../Azure.AI.Extensions.OpenAI
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Projects.Agents\src/../
cd ../Azure.AI.Extensions.OpenAI
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Extensions.OpenAI\src/../
cd ../Azure.AI.Projects
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Projects\src/../
cd <cs_root>/sdk/ai/Azure.AI.Projects.Agents
dotnet build /t:GenerateCode
cd ../Azure.AI.Extensions.OpenAI
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Extensions.OpenAI\src/../
cd ../Azure.AI.Projects
npm exec --prefix <cs_root>\eng\common/tsp-client --no -- tsp-client update --no-prompt --output-dir<cs_root>\sdk\ai\Azure.AI.Projects\src/../
Azure.AI.Projects, Azure.AI.Extensions.OpenAI or Azure.AI.Projects.Agents):cd <cs_root>/sdk/ai/<package>
dotnet build /t:GenerateCode
cd <cs_root>/sdk/ai/<package>
dotnet build /t:GenerateCode
Azure.AI.Projects.Agents package after code generation may get new tool classes, inherited from Tool. The same classes will be generated in Azure.AI.Extensions.OpenAI projects. If that is the case, add into the file <cs_root>/sdk/ai/Azure.AI.Extensions.OpenAI/src/Custom.CodeGenStubs.Tools.cs entries to rename the generated tool classes and their parameters so that they have Responses prefix. For example, if the class FabricIQPreviewTool was added, add the next entries:[CodeGenType("FabricIQPreviewTool")] public partial class ResponsesFabricIQPreviewTool { }
[CodeGenType("FabricIQPreviewToolParameters")] public partial class ResponsesFabricIQPreviewToolParameters { }
Order the entries in the file alphabetically; do not remove the entries already present.
dotnet build inside the project folder.Azure.AI.Projects, Azure.AI.Extensions.OpenAI or Azure.AI.Projects.Agents):cd <cs_root>
eng\scripts\Export-API.ps1 ai\<package>
cd <cs_root>
eng\scripts\Export-API.ps1 ai\<package>
A C# sample consists of two parts: the source code in the folder <cs_root>/sdk/ai/<package_name>/tests/Samples and corresponding .md files go to <cs_root>/sdk/ai/<package_name>/Samples.
The C# file should demonstrate both sync and async API when appropriate. See the example below:
#if SNIPPET
var projectEndpoint = System.Environment.GetEnvironmentVariable("FOUNDRY_PROJECT_ENDPOINT");
var modelDeploymentName = System.Environment.GetEnvironmentVariable("FOUNDRY_MODEL_NAME");
#else
var projectEndpoint = TestEnvironment.FOUNDRY_PROJECT_ENDPOINT;
var modelDeploymentName = TestEnvironment.FOUNDRY_MODEL_NAME;
#endif
AIProjectClient projectClient = new(endpoint: new Uri(projectEndpoint), tokenProvider: new DefaultAzureCredential());
DeclarativeAgentDefinition agentDefinition = new(model: modelDeploymentName)
{
Instructions = "You are a prompt agent."
};
ProjectsAgentVersion agentVersion1 = await projectClient.AgentAdministrationClient.CreateAgentVersionAsync(
agentName: "myAgent1",
options: new(agentDefinition));
#if SNIPPET
var projectEndpoint = System.Environment.GetEnvironmentVariable("FOUNDRY_PROJECT_ENDPOINT");
var modelDeploymentName = System.Environment.GetEnvironmentVariable("FOUNDRY_MODEL_NAME");
#else
var projectEndpoint = TestEnvironment.FOUNDRY_PROJECT_ENDPOINT;
var modelDeploymentName = TestEnvironment.FOUNDRY_MODEL_NAME;
#endif
AIProjectClient projectClient = new(endpoint: new Uri(projectEndpoint), tokenProvider: new DefaultAzureCredential());
DeclarativeAgentDefinition agentDefinition = new(model: modelDeploymentName)
{
Instructions = "You are a prompt agent."
};
ProjectsAgentVersion agentVersion1 = projectClient.AgentAdministrationClient.CreateAgentVersion(
agentName: "myAgent1",
options: new(agentDefinition));
The code needs to be logically split into blocks, whose usage should be explained in .md file. Each block is marked in the source code as follows:
#region Snippet:Sample_FunctionFoo_MySampl_Async
public async Task Foo()
{
await Task.Delay(1000);
Console.WriteLine("Hello world!")
}
#endregion
These blocks will be rendered in the .md file code block based on the region name (it will be done by automation):
public async Task Foo()
{
await Task.Delay(1000);
Console.WriteLine("Hello world!")
}
If the code block has sync and async counterpart, please create code block for each of them. Provide the short explanation for each code block. Please make sure that there is one-to-one consistency between C# regions and .md file code blocks.
To make sure that it is so, call the script (replace <package> by Azure.AI.Projects, Azure.AI.Extensions.OpenAI or Azure.AI.Projects.Agents):
cd <cs_root>
eng\scripts\Update-Snippets.ps1 ai\<package>
cd <cs_root>
eng\scripts\Update-Snippets.ps1 ai\<package>
Get the differences by running
cd <cs_root>
git diff
cd <cs_root>
git diff
Figure out, which classes are public facing and based on that populate the latest section of Release History. Append the found changes to the ### Features Added, ### Breaking Changes, ### Bugs Fixed or ### Other Changes section. It is also possible to add ### Sample Updates section or append data to it. Do not modify the header, denoting version and release date, for example ## 2.1.0-beta.2 (Unreleased).
gh api user --jq .login. git stash
git checkout main
git pull origin main
git stash
git checkout main
git pull origin main
git checkout -b <user_alias>/code_generation_<commit>
git stash apply
git checkout -b <user_alias>/code_generation_<commit>
git stash apply
git add and modified files by git add -u.git commit -m "new features"gh pr create --title "<title>" --body "<changelog> --assignee @me"; replace \<title\> and \<changelog\> by short PR title and the changes from the changelog, we have generated in "Updating changelog" section.az az pipelines show --id 7296 --organization https://dev.azure.com/azure-sdk/ --project internal).curl https://dev.azure.com/azure-sdk/internal/_artifacts/feed/azure-sdk-for-net-pr)az pipelines run --id 7296 --organization https://dev.azure.com/azure-sdk/ --project internal --branch nirovins/fix_pipelines --variables "SetDevVersion=true" --open; replace the \<branch\> by the branch from "Create a pull request" section.After everything is ready, please generate the packages for all \<package\> i.e. (replace \<package\> by Azure.AI.Projects, Azure.AI.Extensions.OpenAI or Azure.AI.Projects.Agents).
cd <cs_root>/sdk/ai/<package>
dotnet pack
cd <cs_root>/sdk/ai/<package>
dotnet pack
Provide the location of each package in the summary and provide the simple instruction on installation.
If the alpha package was released to Azure (in section "Create an alpha package release"), please add to the instructions command to add a new NuGet repository to the project. dotnet nuget add source https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-net/nuget/v3/index.json
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Automatically creates user-facing changelogs from git commits by analyzing commit history, categorizing changes, and transforming technical commits into clear, customer-friendly release notes. Turns hours of manual changelog writing into minutes of automated generation.
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification
Take azure/generate-code-cs from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.