- Detect async/sync generators after tool execution and materialize
into lists before the result conversion pipeline processes them
- Generator materialization runs inside timeout scope so slow generators
respect the configured timeout
- Handle bytes return types: UTF-8 bytes as text, non-UTF-8 as base64
- Suppress output_schema for bytes return types (can't be structured JSON)
- Catch UnicodeDecodeError alongside PydanticSerializationError in
convert_result for robustness
Fixes#3829
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
pydocket 0.19.0 fixes the fakeredis 2.35.0 FakeConnection rename
internally, so we no longer need to carry the fakeredis ceiling
ourselves. Removes the direct fakeredis[lua]<2.35.0 dependency from the
tasks extra entirely — it's just a transitive of pydocket now.
🤖 Generated with Claude Code
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: resolve OpenAPI 3.x server variables in _create_default_client
When an OpenAPI spec defines server variables (e.g. `https://{region}.api.example.com/v1`),
the default values are now substituted before constructing the httpx client base URL.
Previously, the URL was used as-is, causing all requests to fail for specs that use
server variable templating.
Fixes#1681
* fix: use str.replace instead of format_map for server variable substitution
format_map applies Python string formatting rules, so variable names
like {api.version} would be treated as attribute access and raise errors.
Literal token replacement handles all valid OpenAPI variable names safely.
* Unify background task context forwarding and fix concurrent dependency bugs
We've been getting a steady trickle of edge-case reports around background tasks
and contextual dependencies over the last few months (#3654, #3656, #3569). Each
one pointed at a different symptom, but they all traced back to the same area:
the way context is negotiated between the "frontend" server and Docket workers
was grown piecemeal, with each new piece of context (access tokens, HTTP headers,
origin request IDs) getting its own Redis key, its own restore function, and its
own ContextVar. This made it hard to reason about what state was available where,
and the shared-instance Dependency pattern made concurrent tasks stomp on each
other's cleanup state.
This takes a step back and reworks the whole thing as a single unified system:
- Dependency subclasses (_CurrentContext, Progress, _CurrentAccessToken, etc.)
are now stateless factories — __aenter__ returns a fresh per-invocation
object, so concurrent tasks never share mutable state. Fixes#3654, #3656.
- The three individual context-snapshot Redis keys (access_token, http_headers,
origin_request_id) are collapsed into a single TaskContextSnapshot stored as
one JSON key per task. The three _restore_task_* functions and two ContextVars
they populated are gone.
- Sync functions like get_http_request() and get_access_token() now find the
snapshot transparently in background tasks via a 3-tier sync fallback:
ContextVar (set by _CurrentContext for functions with deps) → in-memory dict
(same-process workers) → sync Redis GET (out-of-process workers). No function
wrapping needed.
- The _wrap_for_task_http_headers hack is deleted. FunctionTool registers its
raw function with Docket so Docket sees and resolves ALL dependencies,
including Docket-native ones like Retry and Timeout.
- ProxyTool.from_mcp_tool() now propagates execution.taskSupport metadata from
remote tools. Fixes#3569.
- Removed redundant _current_docket/_current_worker ContextVar management from
Context.__aenter__/__aexit__ (they're only set in the lifespan now).
Closes#3654Closes#3656Closes#3569🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Address code review feedback
- _OptionalCurrentContext: guard __aexit__ against cleaning up contexts it
didn't create (check is_background_task before delegating)
- Narrow except clauses in snapshot loading (OSError, JSONDecodeError, etc.
instead of bare Exception)
- Fix docstrings on register_with_docket for resources/prompts/templates
- Simplify Progress: read ExecutionProgress directly from current_execution
instead of creating and manually entering a DocketProgress wrapper
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Use pop-on-access transfer buffer instead of bounded LRU cache for snapshots
The in-memory snapshot dict is a transfer mechanism, not a cache. Entries go
in at submission and come out at the worker's first access. Using pop instead
of get means the dict only holds entries during the brief submission-to-execution
window, bounded by task concurrency (~10) rather than a 10,000-entry LRU limit.
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Drop in-memory transfer buffer, use sync Redis for all backends
Instead of maintaining an in-memory dict to bridge the async/sync gap, use
a sync Redis client directly. For memory:// backends (fakeredis), shares the
same FakeServer instance via docket._redis.get_memory_server() so data written
by the async Docket client is visible to sync reads. For real Redis, creates a
standard sync connection. No in-process state to manage at all.
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Move snapshot operations to TaskContextSnapshot methods
capture(), from_json(), to_json(), save() are now classmethod/instance methods
on the dataclass instead of free functions. Deduplicates JSON parsing that was
copy-pasted between the async and sync load paths.
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Trim implementation details from register_with_docket docstrings
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Clarify docket lookup comment in submit_to_docket
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Restore docket/worker ContextVar bridge in Context.__aenter__
Servers that own the Docket (the parent) re-set _current_docket/_current_worker
from their instance attributes when entering a Context. Mounted children skip
this (their _docket is None), so they inherit the parent's value. This is needed
for ASGI deployments where ContextVars set during the lifespan don't propagate
to request handlers.
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Key snapshot cache by task_id to prevent cross-task context leakage
Docket workers may reuse the same asyncio context for sequential tasks.
The ContextVar cache now stores (task_id, snapshot) tuples so stale entries
from previous tasks are automatically ignored.
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Transparently refresh upstream token in OAuthProxy.load_access_token()
When upstream token validation fails during load_access_token, attempt
to refresh using the stored refresh token before returning None. This
prevents premature 401s that force clients into expensive full re-auth
flows when the upstream token expires.
Co-authored-by: Claude <noreply@anthropic.com>
* Gate transparent refresh on token expiry, add advisory lock
Only attempt upstream refresh when the token is actually expired, not
on any validation failure (scope mismatch, revocation, etc.). Add
per-token advisory lock to prevent concurrent async tasks from racing
to refresh the same upstream token.
* Re-check expiry inside lock, reload from storage after refresh failure
---------
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add TokenCache utility and caching to GitHubTokenVerifier
Extract the caching machinery from IntrospectionTokenVerifier into a
shared TokenCache class in fastmcp.utilities.token_cache, then wire
it into both IntrospectionTokenVerifier and GitHubTokenVerifier.
* Remove dead constant, validate negative cache params
* Fix overwrite eviction bug, skip cache on scope lookup failure
* pin pydantic-monty to 0.0.8
* rename tool/prompt/resource base modules to avoid decorator name shadow
* add sys.modules shims for old submodule import paths
* preserve original module paths in deprecation warnings
* clarify when sys.modules shims can be removed
* fix: enforce auth/visibility in ResourcesAsTools and PromptsAsTools for non-FastMCP providers
🤖 Co-authored-by: Claude <noreply@anthropic.com>
* fix: honor stdio auth bypass and correct transform ordering in provider wrappers
Co-authored-by: Claude <noreply@anthropic.com>
* fix: move context/dependencies imports into function to break circular import
* fix: route ResourcesAsTools/PromptsAsTools through ctx.fastmcp
Instead of manually reimplementing auth, visibility, and session
transforms in the transform layer, tool functions now call
ctx.fastmcp.read_resource() / ctx.fastmcp.render_prompt() which
routes through the server's full middleware chain. This matches
the pattern CodeMode uses with ctx.fastmcp.call_tool().
The isinstance(provider, FastMCP) branching is removed entirely.
* feat: add _scope parameter for provider-scoped listing
AggregateProvider can now filter which child providers to query when
listing components. ResourcesAsTools and PromptsAsTools use this to
scope listings to their configured provider while still routing
through ctx.fastmcp for full middleware coverage.
The scope matching walks wrapped providers, so a
WrappedProvider(Namespace, inner=MyProvider) matches if MyProvider
is in the scope list.
* test: add coverage for ResourcesAsTools scoped to a sub-server
* fix: delegate to super() when _scope is None, add AggregateProvider to scope matching
* simplify: remove _scope machinery, route everything through ctx.fastmcp
Reverts the _scope parameter from Provider/AggregateProvider/Server.
ResourcesAsTools and PromptsAsTools now simply route through
ctx.fastmcp for all operations. Apply to a FastMCP server instance
for proper auth/visibility/middleware coverage.
Tests rewritten to use FastMCP server directly instead of raw providers.
* warn when ResourcesAsTools/PromptsAsTools is applied to a non-FastMCP provider
* docs: explain that ResourcesAsTools/PromptsAsTools should wrap a FastMCP server
* raise TypeError instead of warning when applied to non-FastMCP provider
---------
Co-authored-by: Claude <noreply@anthropic.com>
Keycloak returns refresh_expires_in=0 for offline tokens (offline_access scope),
meaning "no fixed time-based expiry". The truthiness check on this value caused
the proxy to skip issuing a PROXY_RT, forcing browser re-auth every hour.
Closes#3509🤖 Generated with Claude Code
Co-authored-by: Marvin Context Protocol <41898282+Marvin Context Protocol@users.noreply.github.com>
Co-authored-by: Jeremiah Lowin <jlowin@users.noreply.github.com>
* feat: make upstream_client_secret optional in OAuthProxy
Extract _create_upstream_oauth_client() factory method for subclass
override. Cookie signing falls back to JWT key material when no secret.
* fix: include client_id in revocation requests for public clients
* fix: use factory method for revocation auth
- Use 10 PBKDF2 iterations in test_mode (vs 1M in production) for
JWT key derivation — cuts auth test setup from ~2.5s to <0.1s
- Add timeout(15) to subprocess-spawning tests (TestKeepAlive,
test_mcp_config) that exceed 5s under parallel CI load
- Remove pytestmark filterwarnings overrides in tests/deprecated/
that were leaking DeprecationWarning to test output
- Fix deprecated add_tool_transformation() usage in test_authorization
- Document new settings in settings.mdx
* perf: expose minimum_check_interval, reduce task pickup latency
The Docket Worker polls for new tasks every minimum_check_interval
(previously hardcoded to 250ms in pydocket). Expose this setting so
users can tune it, default to 50ms, and override to 10ms in tests.
This cuts average task pickup latency from ~125ms to ~5ms per task.
* perf: reduce task test overhead and eliminate cross-test contamination
- Expose minimum_check_interval setting (default 50ms, 10ms in tests)
to reduce Docket Worker task pickup latency
- Isolate fakeredis per test via unique memory:// URLs to prevent
stale _async_blocking tasks from contaminating subsequent tests
- Make client disconnect timeout configurable (default 5s, 1s in tests)
- Add --durations=50 to CI for passive performance regression detection
- Remove 15s timeout band-aids from task test conftest files
- Add explicit @pytest.mark.timeout(10) to cancellation tests
- Fix deprecated FastMCP.as_proxy() usage in test_task_proxy.py
* Propagate x-fastmcp-wrap-result flag in tool result _meta
🤖 Generated with Claude Code
Co-authored-by: Claude <noreply@anthropic.com>
* Skip listTools round-trip when _meta has x-fastmcp-wrap-result
🤖 Generated with Claude Code
Co-authored-by: Claude <noreply@anthropic.com>
* Use namespaced meta key: {"fastmcp": {"wrap_result": true}}
🤖 Generated with Claude Code
Co-authored-by: Claude <noreply@anthropic.com>
* Clean up _parse_call_tool_result: hoist cast import, document local CallToolResult import, extract fastmcp_meta
🤖 Generated with Claude Code
Co-authored-by: Claude <noreply@anthropic.com>
* Merge _meta in tasks result handler instead of overwriting
🤖 Generated with Claude Code
* Preserve type validation in meta-based unwrap path
🤖 Generated with Claude Code
* Fix type validation for wrapped task results, guard non-dict meta
🤖 Generated with Claude Code
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix: forward custom_route endpoints from mounted servers
When a child server with custom HTTP routes (registered via
@server.custom_route()) is mounted onto a parent, the routes were
silently dropped because _get_additional_http_routes() only returned
self._additional_http_routes without recursing into mounted providers.
This caused 404s for endpoints like /readyz health checks that worked
in v2 but broke in v3 (regression).
The fix updates _get_additional_http_routes() to traverse providers,
unwrap _WrappedProvider layers (from namespace transforms), find
FastMCPProvider instances, and recursively collect their server's
custom routes.
Fixes#3457
* fix: narrow type annotation from BaseRoute to Route
All items in _additional_http_routes are Route objects (created via
Route(...) in custom_route()). Using list[Route] instead of
list[BaseRoute] fixes the ty type checker failure where .path is
accessed on BaseRoute which doesn't have that attribute.
Removes unused BaseRoute imports from both server.py and transport.py.
* fix: revert route type to list[BaseRoute] to fix ty errors
The previous commit narrowed _additional_http_routes from list[BaseRoute]
to list[Route], which broke:
- component_manager appending Mount objects (Mount is BaseRoute, not Route)
- tests assigning list[BaseRoute] variables (generics are invariant)
Revert to list[BaseRoute] and use isinstance(r, Route) guards in tests
for type-safe .path access.
* fix: remove unused import and fix import grouping
- Remove unused `Route` import from server.py
- Fix import grouping in test_advanced.py (ruff check)
* Address review: move imports to module root, type Provider, add collision note
* fix: sort imports in transport.py
---------
Co-authored-by: voidborne-d <voidborne-d@users.noreply.github.com>
Co-authored-by: d 🔹 <258577966+voidborne-d@users.noreply.github.com>
Co-authored-by: Jeremiah Lowin <153965+jlowin@users.noreply.github.com>
* fix: shield lifespan teardown from cancellation
Co-authored-by: Claude <noreply@anthropic.com>
* fix: stabilize flaky task and timeout tests under parallel execution
Co-authored-by: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix: cache component lists in ProxyProvider to avoid redundant backend connections
Every call_tool through a proxy was triggering _list_tools() to resolve
the tool by name, opening a full MCP session just for the lookup, then
opening a second session for the actual execution. This caches component
lists on the ProxyProvider with a configurable TTL (default 300s),
cutting backend handshakes in half for repeated calls.
* docs: document component caching and session reuse for proxy providers
* fix: add sleep in cache TTL test for Windows clock resolution
* docs: clarify cache scope and dynamic backend guidance
* fix: normalize Google scope shorthands and surface valid_scopes
Google accepts shorthand scopes like "email" in authorization requests but
returns full URIs like "https://www.googleapis.com/auth/userinfo.email" in
token responses. The verifier now normalizes shorthands at initialization so
the subset check works regardless of which form was used. GoogleProvider also
now exposes valid_scopes for controlling which scopes clients can request
beyond the required minimum.
Co-authored-by: Claude <noreply@anthropic.com>
* remove unused GOOGLE_SCOPE_ALIASES_REVERSE
---------
Co-authored-by: Claude <noreply@anthropic.com>
When OIDCProxy has verify_id_token=True and the IdP issues the same JWT
for both access_token and id_token, the value-equality check
`verification_token != upstream_token_set.access_token` evaluated to
False, skipping the scope patch entirely. This left AccessToken.scopes
empty, causing RequireAuthMiddleware to return 403 insufficient_scope.
Replace the value-equality check with an intent-based virtual method
`_uses_alternate_verification()` that OIDCProxy overrides to return
`self._verify_id_token`. The base OAuthProxy returns False (preserving
existing behavior for non-OIDC providers).
Fixes#3461
Co-authored-by: voidborne-d <voidborne-d@users.noreply.github.com>
* Block HS* JWT verification with public keys/JWKS
🤖 Generated with GPT-5.2-Codex
* Fix ruff format violations
🤖 Generated with Claude Code
* Handle bytes public_key in HS* algorithm PEM check
* Fix get_* returning None when latest version is disabled (#3421)
When a visibility transform disabled the highest version of a component,
get_tool/get_resource/get_resource_template/get_prompt returned None
instead of falling back to the next-highest enabled version. The list_*
path already worked correctly because deduplication runs after visibility
filtering. The get_* path now falls back to listing all versions and
picking the highest enabled one when the top version is disabled.
* Apply auth checks in version fallback paths
The fallback code in get_tool, get_resource, get_resource_template, and
get_prompt bypassed auth filtering when falling back to older versions
after the highest version was disabled. This could expose auth-protected
older versions to unauthorized users.
