mcpsmith

Turn an OpenAPI/Swagger file, or a website session you recorded in the browser, into a curated, tested MCP server project. An LLM designs the tools; you get ordinary Python code you own.

Most generators mirror every endpoint as a tool and hand the model raw JSON, or stop at an OpenAPI spec or API client. mcpsmith designs a few workflow-level tools with plain argument names, descriptions that say when to call what next, and compact output. The input is an OpenAPI file or a HAR recording, so sites with no public API work too. The output is a server project with tests, CI, a PyPI + MCP Registry publish workflow and a Claude Code plugin.

Quickstart: OpenAPI

uvx mcpsmith inspect https://petstore3.swagger.io/api/v3/openapi.json   # .mcpsmith/source.json + digest.md
uvx mcpsmith draft --name petstore-mcp                                  # heuristic .mcpsmith/plan.yaml, no LLM
# redesign plan.yaml (your coding agent does this, or use `design`, see below)
uvx mcpsmith validate
uvx mcpsmith generate -o petstore-mcp
cd petstore-mcp && uv run pytest
uvx mcpsmith check .                                                    # call every tool live
claude mcp add petstore-mcp -- uv run --directory "$PWD" petstore-mcp

A spec read from a local file may only have a relative server URL; validate then asks you to set an absolute base_url in the plan.

Quickstart: a website without an API

pip install "mcpsmith[capture]" && playwright install chromium
mcpsmith capture https://example.com -o site.har --host example.com   # browse, then close the tab
mcpsmith inspect site.har --host example.com
mcpsmith draft --name example-mcp
# redesign plan.yaml, then validate, generate, check as above

While recording, do each thing you want as a tool: search, open a detail page, paginate, apply a filter. Anything you do not click is not in the capture. --host (repeatable) keeps only the hosts you list. A DevTools "Save all as HAR" export works in place of capture. The recording browser uses HTTPS_PROXY when it is set.

A HAR holds everything the browser sent and received, including cookies and your personal data: keep it out of git and delete it after inspect. inspect scrubs what it keeps: an API key or token becomes an auth env var with its value dropped, other secret-looking headers and query params are dropped, and emails, phone numbers, JWTs and values under secret or personal field names are masked in examples and test fixtures. This is pattern-based, so read tests/fixtures/ before you publish a generated server. If a key is a public one the site sends to every visitor (a search or maps key), you may put it in the tool's fixed values.

Use it from your agent

No API key needed: your coding agent is the designer.

• Claude Code plugin (skills + MCP server): /plugin marketplace add sepehr071/mcpsmith, then /plugin install mcpsmith@mcpsmith.
• MCP server only, any client:

  {"mcpServers": {"mcpsmith": {"command": "uvx", "args": ["mcpsmith", "mcp"]}}}
  

  Tools: smith_guide, smith_inspect, smith_draft, smith_validate, smith_generate, smith_check (paths absolute), and for the hub smith_record_start, smith_record_stop, smith_connect, smith_disconnect, smith_connections, smith_tools, smith_call. MCP clients start servers without your shell's proxy variables; behind a proxy, add "env": {"HTTPS_PROXY": "http://127.0.0.1:8080"} so smith_inspect can fetch specs.
• Skills: skills/build-mcp-server (the whole workflow), skills/extract-api (logged-in sites into live tools) and skills/design-mcp-tools (design checklist). AGENTS.md points other agents (Codex, Cursor, Gemini CLI) at them.

Then ask: "Build an MCP server for this OpenAPI file; users should be able to find pets and check stock."

Live hub

Skip the project: connect a plan and its tools appear in your agent right away. The mcpsmith MCP server is the hub; connections are saved under ~/.mcpsmith (MCPSMITH_HOME) and load again on restart. They are personal (one machine, one user); generate stays the way to publish.

From the agent: smith_inspect, edit plan.yaml, smith_validate, then smith_connect(workdir, name) and call the new tools. If your client does not show tools added after start, use smith_tools(connection) and smith_call(connection, tool, args) (read-only tools only). smith_connections lists, smith_disconnect(name) removes (CLI: mcpsmith connections).

Credentials come from you, never the chat: mcpsmith secret set <connection> <ENV_NAME> asks with a hidden prompt and stores the value in the OS keychain (--once forgets it after one use, for one-time codes; list and delete also exist).

Logged-in sites

For a dashboard or portal with no public API, the agent records the site's own calls while you use it. The extract-api skill runs the whole flow.

mcpsmith browser                                                      # shared Chrome on port 9222 (--port N, --no-proxy)
npx chrome-devtools-mcp@1.11.0 --browser-url=http://127.0.0.1:9222    # point Chrome DevTools MCP at it

1. Log in yourself in that Chrome window (passwords and 2FA never reach the agent).
2. The agent calls smith_record_start(name, hosts), browses each feature twice with different inputs, then smith_record_stop(name), which writes a HAR with cookies and tokens dropped.
3. It inspects, designs and connects the plan with auth: {type: browser, login_path: /login}. Calls reuse your Chrome session; when it expires the tool says to log in again.

Agents must not use DevTools get_network_request: it prints request headers, including cookies. Plans convert formats with transform on arguments, output.transforms (date->unix, date->unix_end, unix->date, unix->datetime, minutes->hh:mm, toman->rial, rial->toman, int->str, plus optional Jalali calendar conversions) and timezone (IANA name; empty means the system's local zone). generate refuses browser-auth plans because they only work with your Chrome; switch to token or API-key auth to publish.

Let mcpsmith call an LLM

uvx mcpsmith design --goal "find products, compare prices, read reviews" --provider anthropic

Providers: anthropic (ANTHROPIC_API_KEY), openai (OPENAI_API_KEY), openrouter (OPENROUTER_API_KEY), ollama (no key). --model picks the model; MCPSMITH_MODEL and MCPSMITH_BASE_URL override the model and endpoint. The result is the same plan.yaml; review it before generate.

How it works

openapi.yaml | site.har --inspect--> source.json + digest.md
                        --draft----> plan.yaml     (heuristic starting point)
   your agent edits plan.yaml, or `design` ------> plan.yaml
                        --validate-> issues
                        --generate-> <name>/       (server project)
                        --check----> live report per tool

Source is what the API offers; Plan is the server to build; code generation reads the plan plus captured responses for test fixtures. A plan, trimmed:

name: shop-mcp
description: Search products and compare prices on Example Shop.
prefix: shop
base_url: https://example.com/api
env_prefix: SHOP_MCP
instructions: Search first, then fetch details with the id. Prices are in toman.
tools:
  - name: shop_search
    title: Search products
    description: Returns matching products with price and stock. Use shop_product for full details.
    endpoint: GET /v2/search
    args:
      - {name: query, type: string, required: true, description: "Search words, e.g. 'laptop'", to: query.q}
      - {name: max_price, type: integer, description: "Upper price limit in toman, e.g. 20000000", to: query.price_max}
      - {name: sort, description: "Result order.", map: {cheapest: price_asc, newest: date_desc}, default: cheapest, to: query.o}
    output:
      root: data.products
      fields: {id: id, title: title_fa, price: price.selling_price, in_stock: stock.available}
      max_items: 10
    example: {query: laptop}

Argument targets (to:): path.<n>, query.<n>, header.<n>, cookie.<n>, body, body.<a.b.c>. Output paths use dots, a number indexes a list (images.0.url), [] maps over a list (tags[].name). map turns friendly values the model sees into the cryptic ones the API wants.

What you get

<name>/
  pyproject.toml  server.json  README.md  LICENSE  .gitignore  mcpsmith.yaml
  src/<package>/server.py        # tools, built on mcpsmith.runtime
  tests/                         # replay captured responses; one -m live test per tool
  .github/workflows/             # ci.yml, publish.yml (PyPI + MCP Registry)
  .claude-plugin/plugin.json     # Claude Code plugin
  skills/<name>/SKILL.md         # skill for the generated server

mcpsmith.yaml is the plan, so you can regenerate. To publish: add a PyPI trusted publisher (workflow publish.yml, environment pypi), bump the version in pyproject.toml and server.json, tag v*.

Runtime features

Generated servers import mcpsmith.runtime, so a fix there reaches every server on upgrade.

• Retries on network errors, timeouts, 429 and 5xx (writes only on 429, so they are never sent twice); Retry-After honoured and capped; jitter; concurrency cap.
• GET response cache, host allowlist, 5 MB body cap.
• Proxy support: <P>_PROXY takes one URL or a comma list (pool). <P>_BASE_URL overrides the base URL. System proxy env vars are ignored on purpose.
• Chrome TLS fingerprint for sites that block plain clients: pip install "mcpsmith[stealth]" and impersonate: true in the plan. Without the extra it falls back to plain httpx with one warning.
• Compact output: shape() keeps only the planned fields, caps lists (reports the total count), clips long strings, drops nulls.
• Actionable errors: a 403 suggests setting <P>_PROXY, a 401 names the auth env var, 404/429/5xx say what to do next.
• Auth: bearer, header, query, cookie or basic, read from an env var.

Drift check

mcpsmith check <project> [--json] calls every tool with its example arguments and reports errors, output size in characters and output fields that came back empty. Run it nightly in CI to see when an API or site changes. It makes real calls, so tools marked read_only: false are skipped unless you pass --writes.

Comparison

Compiled from each project's public docs in October 2026; check them before relying on a cell. FastMCP's own docs say curated servers clearly beat auto-converted ones.

Limitations

• No browser per call: JS challenges and captchas are out of scope. Sites that need them will fail at check.
• A tool maps to one request. Multi-call workflow tools need hand edits to the generated server.py, and regenerating overwrites them.
• The generated project is Python only.
• A HAR only contains what you did while recording; unrecorded features are not in the plan.
• Request bodies are sent as JSON; endpoints that only accept application/x-www-form-urlencoded or multipart need a hand edit.
• The plan is trusted input: its text ends up in server instructions and the skill, and mcpsmith check runs the generated code. Review a plan written by an LLM from third-party content before you generate and publish it.
• Alpha (0.1).

Develop

uv sync
uv run pytest
uv run ruff check .

License

MIT