本文へ移動
cccskills
無料GitHub で公開

dotnet-mcp-builder

Build Model Context Protocol (MCP) servers in C#/.NET against the current ModelContextProtocol 2.x NuGet packages. Helps with cases the model gets wrong without guidance — stale versions (0.x preview or 1.x-era defaults), the v2 stateless-by-default HTTP flip, the 2026-07-28 spec deprecations (roots/sampling/logging), MCP Apps and Tasks extension packages, elicitation URL mode, per-session HTTP wiring, OAuth and reverse-proxy deploy specifics, and debugging MapMcp / STDIO / Streamable-HTTP errors. Also covers STDIO and Streamable HTTP transports (SSE is deprecated), tools, prompts, resources, completions, and a basic .NET MCP client. Trigger when the user says or implies any .NET MCP server work: ModelContextProtocol, McpServerTool, MapMcp, WithStdioServerTransport, "MCP server in C#", "MCP tool in dotnet", "expose this as MCP", or names a primitive (prompt/resource/elicitation/MCP App) in a .NET context. Skip for MCP work in other languages.

インストール方法を見る

含まれるファイル(14)

  • SKILL.md8.4 KB
  • references/client.md6.2 KB
  • references/elicitation.md7.8 KB
  • references/mcp-apps.md9.8 KB
  • references/packages.md5.2 KB
  • references/prompt-primitive.md5.0 KB
  • references/resource-primitive.md6.1 KB
  • references/roots.md4.0 KB
  • references/sampling.md5.4 KB
  • references/server-features.md6.6 KB
  • references/testing.md6.1 KB
  • references/tool-primitive.md8.4 KB
  • references/transport-http.md8.4 KB
  • references/transport-stdio.md5.1 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Building MCP servers in .NET

This skill helps you write production-quality MCP servers and basic clients in C#/.NET against the official ModelContextProtocol NuGet packages, maintained by Microsoft and the MCP project. It targets the stable 2.x line and the current spec (2026-07-28).

When this skill earns its keep

The .NET MCP SDK had years of preview packages (0.x-preview) before reaching 1.0, and v2 flipped several defaults. Without help, the model tends to:

  • Pin a stale preview version that won't compile against current samples.
  • Apply 1.x-era defaults that v2 reversed (HTTP was stateful by default; in 2.x Stateless defaults to true).
  • Recommend capabilities the 2026-07-28 spec deprecates (roots, sampling, MCP-channel logging — now [Obsolete], warning MCP9005).
  • Miss recent spec features (multi-round-trip input_required, discovery-first negotiation, MCP Apps/Tasks extension packages, elicitation URL mode, structured content blocks).
  • Get HTTP transport details wrong (stateful/stateless, proxy buffering, OAuth wiring).
  • Forget the STDIO stdout/stderr trap.

If the task is one of those, load the matching reference and follow it. If it's truly trivial (e.g. "rename this tool method"), you don't need to read everything — the cardinal rules below are the minimum.

Mental model in 30 seconds

A .NET MCP server is an ordinary Microsoft.Extensions.Hosting (or WebApplication) app that wires an MCP server through DI:

builder.Services
    .AddMcpServer()
    .WithStdioServerTransport()      // OR .WithHttpTransport(...)
    .WithToolsFromAssembly()         // discover [McpServerToolType] classes
    .WithPrompts<MyPrompts>()        // optional
    .WithResources<MyResources>();   // optional

Primitives are plain C# methods on classes marked with attributes ([McpServerToolType] + [McpServerTool], [McpServerPromptType] + [McpServerPrompt], [McpServerResourceType] + [McpServerResource]). Parameters bind from JSON-RPC; the SDK builds the JSON Schema from the signature plus [Description] attributes.

Server-to-client features (elicitation, progress notifications, and the now-deprecated sampling/roots/log notifications) are methods on the injected IMcpServer.

Decision tree → which references to load

Always load references/packages.md if you're creating a new project or unsure of the current package version.

TaskLoad
New STDIO serverreferences/transport-stdio.md
New HTTP (Streamable) serverreferences/transport-http.md
Add/modify a toolreferences/tool-primitive.md
Add/modify a promptreferences/prompt-primitive.md
Add/modify a resourcereferences/resource-primitive.md
Ask the user a question mid-toolreferences/elicitation.md
Call the client's LLM from a tool (deprecated in 2026-07-28)references/sampling.md
Read the user's project roots (deprecated in 2026-07-28)references/roots.md
Return an interactive UIreferences/mcp-apps.md
Argument completions, log/progress notifications, filters, server instructionsreferences/server-features.md
Write a .NET program that consumes an MCP serverreferences/client.md
MCP Inspector, in-memory tests, mocks, CIreferences/testing.md

For multi-primitive tasks, load several at once. For trivial edits in an existing file, you usually don't need any.

Cardinal rules (apply always; these prevent the highest-frequency breakages)

  1. Pin the current stable package, not a preview. Use ModelContextProtocol / ModelContextProtocol.AspNetCore / ModelContextProtocol.Core at the latest 2.x. If you find yourself writing 0.3-preview or 0.4-preview, stop and check NuGet — preview APIs have breaking differences. 1.x still works but predates the 2026-07-28 spec.
  2. STDIO servers must not write to stdout. Stdout is the JSON-RPC channel. Configure LogToStandardErrorThreshold = LogLevel.Trace before anything else and never Console.WriteLine from a tool.
  3. HTTP defaults to stateless in 2.x (v1.x defaulted to stateful — the single most impactful v2 breaking change). The 2026-07-28 revision has no HTTP sessions at all: setting Stateless = false makes the server refuse that revision and serve clients via the legacy initialize fallback. For "ask the user something mid-tool" on current-protocol HTTP, use the multi-round-trip InputRequiredException pattern; reserve stateful HTTP (or STDIO) for the legacy ElicitAsync/sampling/roots paths and pushed notifications.
  4. SSE-only is deprecated. Use Streamable HTTP. Only enable legacy SSE (EnableLegacySse = true) for an old client you must support, and call it out.
  5. Don't design new servers around deprecated capabilities. The 2026-07-28 spec deprecates roots, sampling, and MCP-channel logging; the SDK marks them [Obsolete] (warning MCP9005). They still work against down-level clients, but for new designs prefer the multi-round-trip input_required pattern and ILogger logging. Suppress MCP9005 only as a documented transition measure.
  6. Always [Description] tools and parameters. This is what the LLM sees when picking and shaping calls. Vague descriptions are the #1 reason tools don't get used.
  7. Show the registration line every time you add a primitive. A new [McpServerPromptType] class without .WithPrompts<...>() (or .WithPromptsFromAssembly()) is invisible.
  8. Don't invent APIs. If you're unsure a method exists, say so and check the API reference — wrong method names cause silent failures. This applies doubly to the new v2 extension packages (ModelContextProtocol.Extensions.Tasks, ModelContextProtocol.Extensions.Apps) — check their docs before writing code against them.

Working style

  • Make minimal, additive changes. Add a method to the existing tool class rather than restructuring the project.
  • For non-trivial setups, run dotnet build. Catches missing usings, attribute typos, and TFM mismatches before the user sees them.
  • Confirm transport + .NET version + primitives before scaffolding if context doesn't already make them obvious. Default to .NET 10 for new projects.

When the user is stuck

Walk this checklist before guessing:

  1. STDIO: something is writing to stdout (logger sink, Console.WriteLine, library banner).
  2. HTTP 404: path mismatch — app.MapMcp() is root, app.MapMcp("/mcp") puts it under /mcp.
  3. Tool not appearing: missing [McpServerToolType] on the class, or no .WithToolsFromAssembly() / .WithTools<T>() registered.
  4. Args not bound: parameter names must match the JSON-RPC arguments keys; complex types bind via System.Text.Json.
  5. Sampling/elicitation/roots failing: these legacy server-to-client calls can't run on current-protocol HTTP — migrate to the multi-round-trip InputRequiredException pattern, or (legacy paths only) set Stateless = false, knowing that pins HTTP clients to a down-level initialize revision. Also check the client actually advertises the capability.
  6. MCP9005 build warnings after upgrading to 2.x: the code uses deprecated roots/sampling/logging APIs. Plan the migration; suppress only temporarily.

Still stuck? Point the user at the EverythingServer sample — it exercises every feature.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Use this skill when the user explicitly asks to map, document, or onboard into an existing codebase. Trigger for prompts like "map this codebase", "document this architecture", "onboard me to this repo", or "create codebase docs". Do not trigger for routine feature implementation, bug fixes, or narrow code edits unless the user asks for repository-level discovery.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

Run the AgentRC readiness assessment on the current repository and produce a static HTML dashboard at reports/index.html. Wraps `npx github:microsoft/agentrc readiness` and hands off rendering to the @ai-readiness-reporter custom agent. Supports policies (--policy) for org-specific scoring. Use when asked to assess, audit, or score the AI readiness of a repo.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

Generate tailored AI agent instruction files via AgentRC instructions command. Produces .github/copilot-instructions.md (default, recommended for Copilot in VS Code) plus optional per-area .instructions.md files with applyTo globs for monorepos. Use after running /acreadiness-assess to close gaps in the AI Tooling pillar.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

Use this skill when the user shares ad campaign performance data and asks what to cut, scale, or test. Trigger for prompts like "analyze my ad campaigns", "where am I wasting ad spend", "reallocate my ad budget", "which ads are actually working", or "ROAS analysis". Do not trigger for campaign planning or creative generation without performance data.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

Add educational comments to the file specified, or prompt asking for file to comment if one is not provided.

日本語の概要は準備中です。原文の説明を表示しています。

github/awesome-copilot4万2026年10月9日 更新

github のスキルをすべて見る

このスキルの問題を報告する