Files
modelcontextprotocol--pytho…/examples/stories/tools/README.md
T
wehub-resource-sync 49b9bb6724
Deploy Docs / deploy-docs (push) Failing after 1s
Conformance Tests / client-conformance (push) Failing after 3s
Conformance Tests / server-conformance (push) Failing after 1s
GitHub Actions Security Analysis / zizmor (push) Failing after 1s
CI / checks (push) Failing after 59m20s
CI / all-green (push) Waiting to run
chore: import upstream snapshot with attribution
2026-07-13 12:10:27 +08:00

40 lines
1.5 KiB
Markdown

# tools
**Start here.** Register tools with `@mcp.tool()`; the SDK infers the JSON
input schema from type hints, the output schema from the return annotation, and
returns `structuredContent` alongside text. `ToolAnnotations` carries
behavioural hints (`readOnlyHint`, `idempotentHint`) the host can show to
users. The client lists tools, inspects schemas + annotations, calls both, and
asserts structured output.
## Run it
```bash
# stdio (default — the client spawns the server as a subprocess)
uv run python -m stories.tools.client
# HTTP — the client self-hosts the server on a free port, runs, then tears it down
uv run python -m stories.tools.client --http
# same, against the lowlevel-API server variant
uv run python -m stories.tools.client --http --server server_lowlevel
```
## What to look at
- `server.py` `calc``Literal[...]` and `BaseModel` in the signature become
the tool's `inputSchema` / `outputSchema` with zero hand-written JSON.
- `server.py` `echo``structured_output=False` opts out of schema inference
for a plain text-only tool.
- `server_lowlevel.py` — the same wire contract built by hand: this is what
`MCPServer` generates for you.
## Spec
[Tools — server features](https://modelcontextprotocol.io/specification/2025-11-25/server/tools)
## See also
`schema_validators/` (every input-schema source: pydantic / TypedDict /
dataclass / dict), `error_handling/` (`is_error` vs protocol error),
`streaming/` (progress mid-call).