Max Isbey
d99b1400c6
The twelve v1 @server.* decorator kinds are gone on v2. Their sites now become add_request_handler / add_notification_handler calls at the decorator's exact source position (registration there is when the v1 decorator ran, so execution order is preserved and the deprecated capabilities land on the warning-free path), wired through generated adapters that reproduce the v1 wrapper semantics: bare-list wrapping, call_tool's any-exception-to-isError contract with jsonschema input and output validation (tool lookup through the registered tools/list handler, v1's own cache mechanism, so cross-module list_tools works), read_resource content conversion, and the completion None-mapping. Handler bodies are never touched. Shapes the adapter cannot serve honestly -- a stacked decorator, an attribute receiver, a non-v1 signature, a non-literal decorator argument, a taken name -- are marked with the reason. The suite migrates a six-registration server and serves it to a v1-shaped ClientSession over the legacy protocol; the templates are pinned against the installed v2 (method strings register, params models exist, imports resolve, no 2026-era surface is emitted). Also on the client surface: inline timedelta session timeouts convert to float seconds and non-provable values are marked (the mismatch only fails on the first request); cursor= on session list_* methods wraps into params=PaginatedRequestParams(...); pydantic URL wrappers around resource URIs are dropped where the target provably takes v2's plain str and marked elsewhere; constructions of and pydantic method calls on the v1 RootModel wrappers that became plain union aliases are marked with the TypeAdapter fix; ._mcp_server and the type-keyed handler dicts are marked with their v2 homes. Adapters honor an explicit `uri: str` annotation and keep v1's AnyUrl otherwise, and keep the emitted code insensitive to user return annotations so a wrong annotation cannot manufacture type errors inside generated code. Batch harness: seven more pinned repositories (two seven-decorator servers, a multi-package lowlevel server, the method-local-server marker path, two client libraries including a positional timedelta timeout and the old streamablehttp spelling, and an exact ==1.6.0 pin). Markers now cover the full statement they precede rather than a fixed radius, Unknown-typed errors in files that carry markers classify as cascade of a marked break, and the work directory is a dot-directory so pytest never collects the cloned repositories' own suites. All eleven repositories audit at zero uncovered errors. An adversarial review round over the full change confirmed ten defects, all fixed with regression tests: adapter imports now inject at the top of the module (a mid-file import as the anchor left registration code running before its imports bound); the rewrite gates now also block a handler named like a template local, and any module-level non-import binding of a name the adapter references (both were silent runtime breaks past the gates); import injection dedup now reads the updated module's top-level import binds, so conditional or function-local imports no longer suppress a needed injection; list_* adapters pass a returned full result model through instead of double-wrapping (v1's runtime behavior); the blocked-progress marker names add_notification_handler (a request-handler registration would never fire); the timeout transform skips already-v2 shapes so re-runs stay no-ops; the emitted name scheme is defined once and shared between templates and gates; and the harness classifier no longer lets a marker cover a whole def/class body or write off arbitrary Unknown-typed errors (header-only spans; cascade restricted to propagation rules and never detonators).
Codemod batch test
Runs the mcp-codemod v1 -> v2 migration against real, pinned repositories and
audits the result with pyright, to find silent misses the unit tests and the
in-repo example corpus cannot.
How it works
For each repository in repos.json:
- Clone the pinned commit (shallow).
- Run the codemod (sources and dependency files) over a copy.
- Type-check the pristine clone against an environment holding the latest v1 SDK, and the migrated copy against this workspace's v2 environment, with identical pyright settings.
- Diff the two error sets. Errors only on the migrated side are the migration surface; baseline noise (the repo's own issues, missing third-party stubs) appears on both sides and cancels out.
- Correlate each new error with the inserted
# mcp-codemod:markers.
The codemod's contract is that the markers are the complete list of remaining
manual work, so every new error should sit on or next to a marker. A new
error with no nearby marker is a silent miss -- those are printed, written to
work/results/<slug>.json, and make the run exit 1.
Usage
From the repository root (the v1 environment is created on first run):
uv run --frozen python scripts/codemod-batch-test/run.py # all repos
uv run --frozen python scripts/codemod-batch-test/run.py --repo mcp-obsidian
uv run --frozen python scripts/codemod-batch-test/run.py --fresh # re-clone
Adding a repository
Add an entry to repos.json with a pinned sha (never a branch), an
include list when only part of the repository uses the SDK (empty means the
whole tree), and a one-line note. Prefer repositories that depend on the
mcp package directly; servers built on the external FastMCP library exercise
that library's surface, not this SDK's.