| name | build-mcp-server |
| description | This skill should be used when the user asks to "build an MCP server", "create an MCP", "make an MCP integration", "wrap an API for Claude", "expose tools to Claude", "make an MCP app", or discusses building something with the Model Context Protocol. It is the entry point for MCP server development โ it interrogates the user about their use case, determines the right deployment model (remote HTTP, MCPB, local stdio), picks a tool-design pattern, and hands off to specialized skills. |
| version | 0.1.0 |
Build an MCP Server
You are guiding a developer through designing and building an MCP server that works seamlessly with Claude. MCP servers come in many forms โ picking the wrong shape early causes painful rewrites later. Your first job is discovery, not code.
Load Claude-specific context first. The MCP spec is generic; Claude has additional auth types, review criteria, and limits. Before answering questions or scaffolding, fetch https://claude.com/docs/llms-full.txt (the full export of the Claude connector docs) so your guidance reflects Claude's actual constraints.
Do not start scaffolding until you have answers to the questions in Phase 1. If the user's opening message already answers them, acknowledge that and skip straight to the recommendation.
Phase 1 โ Interrogate the use case
Ask these questions conversationally (batch them into one message, don't interrogate one-at-a-time). Adapt wording to what the user has already told you.
1. What does it connect to?
| If it connects toโฆ | Likely direction |
|---|
| A cloud API (SaaS, REST, GraphQL) | Remote HTTP server |
| A local process, filesystem, or desktop app | MCPB or local stdio |
| Hardware, OS-level APIs, or user-specific state | MCPB |
| Nothing external โ pure logic / computation | Either โ default to remote |
2. Who will use it?
- Just me / my team, on our machines โ Local stdio is acceptable (easiest to prototype)
- Anyone who installs it โ Remote HTTP (strongly preferred) or MCPB (if it must be local)
- Users of Claude desktop who want UI widgets โ MCP app (remote or MCPB)
3. How many distinct actions does it expose?
This determines the tool-design pattern โ see Phase 3.
- Under ~15 actions โ one tool per action
- Dozens to hundreds of actions (e.g. wrapping a large API surface) โ search + execute pattern
4. Does a tool need mid-call user input or rich display?
- Simple structured input (pick from list, enter a value, confirm) โ Elicitation โ spec-native, zero UI code. Host support is rolling out (Claude Code โฅ2.1.76) โ always pair with a capability check and fallback. See
references/elicitation.md.
- Rich/visual UI (charts, custom pickers with search, live dashboards) โ MCP app widgets โ iframe-based, needs
@modelcontextprotocol/ext-apps. See build-mcp-app skill.
- Neither โ plain tool returning text/JSON.
5. What auth does the upstream service use?
- None / API key โ straightforward
- OAuth 2.0 โ you'll need a remote server with CIMD (preferred) or DCR support; see
references/auth.md
Phase 2 โ Recommend a deployment model
Based on the answers, recommend one path. Be opinionated. The ranked options:
โญ Remote streamable-HTTP MCP server (default recommendation)
A hosted service speaking MCP over streamable HTTP. This is the recommended path for anything wrapping a cloud API.
Why it wins:
- Zero install friction โ users add a URL, done
- One deployment serves all users; you control upgrades
- OAuth flows work properly (the server can handle redirects, DCR, token storage)
- Works across Claude desktop, Claude Code, Claude.ai, and third-party MCP hosts
Choose this unless the server must touch the user's local machine.
โ Fastest deploy: Cloudflare Workers โ references/deploy-cloudflare-workers.md (zero to live URL in two commands)
โ Portable Node/Python: references/remote-http-scaffold.md (Express or FastMCP, runs on any host)
Elicitation (structured input, no UI build)
If a tool just needs the user to confirm, pick an option, or fill a short form, elicitation does it with zero UI code. The server sends a flat JSON schema; the host renders a native form. Spec-native, no extra packages.
Caveat: Host support is new (Claude Code shipped it in v2.1.76; Desktop unconfirmed). The SDK throws if the client doesn't advertise the capability. Always check clientCapabilities.elicitation first and have a fallback โ see references/elicitation.md for the canonical pattern. This is the right spec-correct approach; host coverage will catch up.
Escalate to build-mcp-app widgets when you need: nested/complex data, scrollable/searchable lists, visual previews, live updates.
MCP app (remote HTTP + interactive UI)
Same as above, plus UI resources โ interactive widgets rendered in chat. Rich pickers with search, charts, live dashboards, visual previews. Built once, renders in Claude and ChatGPT.
Choose this when elicitation's flat-form constraints don't fit โ you need custom layout, large searchable lists, visual content, or live updates.
Usually remote, but can be shipped as MCPB if the UI needs to drive a local app.
โ Hand off to the build-mcp-app skill.
MCPB (bundled local server)
A local MCP server packaged with its runtime so users don't need Node/Python installed. The sanctioned way to ship local servers.
Choose this when the server must run on the user's machine โ it reads local files, drives a desktop app, talks to localhost services, or needs OS-level access.
โ Hand off to the build-mcpb skill.
Local stdio (npx / uvx) โ not recommended for distribution
A script launched via npx / uvx on the user's machine. Fine for personal tools and prototypes. Painful to distribute: users need the right runtime, you can't push updates, and the only distribution channel is Claude Code plugins.
Recommend this only as a stepping stone. If the user insists, scaffold it but note the MCPB upgrade path.
Phase 3 โ Pick a tool-design pattern
Every MCP server exposes tools. How you carve them matters more than most people expect โ tool schemas land directly in Claude's context window.
Pattern A: One tool per action (small surface)
When the action space is small (< ~15 operations), give each a dedicated tool with a tight description and schema.
create_issue โ Create a new issue. Params: title, body, labels[]
update_issue โ Update an existing issue. Params: id, title?, body?, state?
search_issues โ Search issues by query string. Params: query, limit?
add_comment โ Add a comment to an issue. Params: issue_id, body
Why it works: Claude reads the tool list once and knows exactly what's possible. No discovery round-trips. Each tool's schema validates inputs precisely.
Especially good when one or more tools ship an interactive widget (MCP app) โ each widget binds naturally to one tool.
Pattern B: Search + execute (large surface)
When wrapping a large API (dozens to hundreds of endpoints), listing every operation as a tool floods the context window and degrades model performance. Instead, expose two tools:
search_actions โ Given a natural-language intent, return matching actions
with their IDs, descriptions, and parameter schemas.
execute_action โ Run an action by ID with a params object.
The server holds the full catalog internally. Claude searches, picks, executes. Context stays lean.
Hybrid: Promote the 3โ5 most-used actions to dedicated tools, keep the long tail behind search/execute.
โ See references/tool-design.md for schema examples and description-writing guidance.
Phase 4 โ Pick a framework
Recommend one of these two. Others exist but these have the best MCP-spec coverage and Claude compatibility.
| Framework | Language | Use when |
|---|
Official TypeScript SDK (@modelcontextprotocol/sdk) | TS/JS | Default choice. Best spec coverage, first to get new features. |
FastMCP 3.x (fastmcp on PyPI) | Python | User prefers Python, or wrapping a Python library. Decorator-based, very low boilerplate. This is jlowin's package โ not the frozen FastMCP 1.0 bundled in the official mcp SDK. |
If the user already has a language/stack in mind, go with it โ both produce identical wire protocol.
Phase 5 โ Scaffold and hand off
Once you've settled the four decisions (deployment model, tool pattern, framework, auth), do one of:
- Remote HTTP, no UI โ Scaffold inline using
references/remote-http-scaffold.md (portable) or references/deploy-cloudflare-workers.md (fastest deploy). This skill can finish the job.
- MCP app (UI widgets) โ Summarize the decisions so far, then load the
build-mcp-app skill.
- MCPB (bundled local) โ Summarize the decisions so far, then load the
build-mcpb skill.
- Local stdio prototype โ Scaffold inline (simplest case), flag the MCPB upgrade path.
When handing off, restate the design brief in one paragraph so the next skill doesn't re-ask.
Beyond tools โ the other primitives
Tools are one of three server primitives. Most servers start with tools and never need the others, but knowing they exist prevents reinventing wheels:
| Primitive | Who triggers it | Use when |
|---|
| Resources | Host app (not Claude) | Exposing docs/files/data as browsable context |
| Prompts | User (slash command) | Canned workflows ("/summarize-thread") |
| Elicitation | Server, mid-tool | Asking user for input without building UI |
| Sampling | Server, mid-tool | Need LLM inference in your tool logic |
โ references/resources-and-prompts.md, references/elicitation.md, references/server-capabilities.md
Phase 6 โ Test in Claude and publish
Once the server runs:
- Test against real Claude by adding the server URL as a custom connector at Settings โ Connectors (use a Cloudflare tunnel for local servers). Claude identifies itself with
clientInfo.name: "claude-ai" on initialize. โ https://claude.com/docs/connectors/building/testing
- Run the pre-submission checklist โ read/write tool split, required annotations, name limits, prompt-injection rules. โ https://claude.com/docs/connectors/building/review-criteria
- Submit to the Anthropic Directory. โ https://claude.com/docs/connectors/building/submission
- Recommend shipping a plugin that wraps this MCP with skills โ most partners ship both. โ https://claude.com/docs/connectors/building/what-to-build
Quick reference: decision matrix
| Scenario | Deployment | Tool pattern |
|---|
| Wrap a small SaaS API | Remote HTTP | One-per-action |
| Wrap a large SaaS API (50+ endpoints) | Remote HTTP | Search + execute |
| SaaS API with rich forms / pickers | MCP app (remote) | One-per-action |
| Drive a local desktop app | MCPB | One-per-action |
| Local desktop app with in-chat UI | MCP app (MCPB) | One-per-action |
| Read/write local filesystem | MCPB | Depends on surface |
| Personal prototype | Local stdio | Whatever's fastest |
Reference files
references/remote-http-scaffold.md โ minimal remote server in TS SDK and FastMCPreferences/deploy-cloudflare-workers.md โ fastest deploy path (Workers-native scaffold)references/tool-design.md โ writing tool descriptions and schemas Claude understands wellreferences/auth.md โ OAuth, CIMD, DCR, token storage patternsreferences/resources-and-prompts.md โ the two non-tool primitivesreferences/elicitation.md โ spec-native user input mid-tool (capability check + fallback)references/server-capabilities.md โ instructions, sampling, roots, logging, progress, cancellationreferences/versions.md โ version-sensitive claims ledger (check when updating)
