- Lead with opentelemetry-instrument as the default approach
- Move programmatic configuration lower in the page
- Remove unimplemented metrics section
- Fix attribute values (resource_template not template)
- Add auth attributes (enduser.id, enduser.scope)
- Add provider-specific delegation attributes
- Link to OpenTelemetry Python docs
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Adds opt-in distributed tracing via OpenTelemetry for observability into
FastMCP server and client operations.
Server spans are created for tool calls, resource reads, and prompt
renders with attributes like component key, component type, provider
type, session ID, and auth context. Client spans wrap outgoing calls
with trace context propagation via W3C headers in request meta.
Components provide their own span attributes through a `get_span_attributes()`
method that subclasses override - this lets LocalProvider, FastMCPProvider,
and ProxyProvider each include relevant context (original names, backend URIs).
To enable: configure an OpenTelemetry SDK with a TracerProvider before
importing fastmcp. Traces export to any OTLP-compatible backend.
Closes ENG-2813
🤖 Generated with Claude Code
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Consolidate tool transformation logic into TransformingProvider
Tool transformations were previously scattered across LocalProvider,
ProxyProvider, and MCPConfig. This consolidates all transformation
logic into TransformingProvider via with_transforms(tool_transforms={...}).
- Add tool_transforms parameter to TransformingProvider
- Add tool_transforms to Provider.with_transforms()
- Remove transformation storage from LocalProvider and ProxyProvider
- Remove add_tool_transformation() and remove_tool_transformation() from FastMCP
- Add tool_transforms parameter to factory methods (from_openapi, from_fastapi, create_proxy)
- Update tests to use new patterns
* Fix: reject tool lookups by pre-transform name
* Add collision validation for tool_transforms and fix docstring examples
- Validate duplicate target names in tool_transforms raise ValueError
- Fix docstring examples to use arguments/ArgTransformConfig (not args/ArgTransform)
- Add test for collision validation
* Add server-level tool transform APIs and fix task registration
- Add AggregateProvider to present multiple providers as one
- Add _get_root_provider() to apply server-level transforms uniformly
- Fix _docket_lifespan to use root provider (ensures renamed tools
register with correct keys for background execution)
- Add tool_transforms kwarg to __init__ (non-deprecated)
- Add add_tool_transform(), remove_tool_transform(), tool_transforms property
- Deprecate old API names (tool_transformations, add_tool_transformation, etc.)
- Update tests to use new API
* Add graceful degradation for provider errors in AggregateProvider
* Match original behavior: parallel queries with DEBUG logging
* Refactor transforms to middleware-style call_next pattern
Replaces the ad-hoc transformation system with a unified Transform
abstraction using the same call_next pattern as server middleware.
Key changes:
- New src/fastmcp/server/transforms/ module with Transform base class
- Namespace, ToolTransform, Visibility all implement the same interface
- Transforms compose via functools.partial chain building
- Visibility is now just the first transform in provider._transforms
- Server-level transforms apply after provider aggregation
- Task registration now applies full transform chain
Removes TransformingProvider, _BoundTransform, ComponentSource protocol.
User-facing API unchanged: mount(), add_transform(), enable/disable all
work as before.
* Add comprehensive transforms and visibility documentation
New docs/servers/providers/transforms.mdx covering:
- Mental model for middleware-style transform pattern
- Built-in transforms (Namespace, ToolTransform)
- Server vs provider-level transforms and ordering
- Tool modification (immediate vs deferred)
- Custom transform creation
New docs/servers/visibility.mdx covering:
- Enable/disable API for runtime visibility control
- Keys and tags for targeting components
- Allowlist mode with only=True
- Server vs provider visibility layering
Updates existing docs to reference new pages and simplifies
redundant content. Visibility is documented as a user feature,
not as an implementation detail.
* Restructure transforms docs and delete tool-transformation pattern
* Cleanup: simplify get_tasks and remove unused Provider.get_component
* Update loq
* Update loq limits and add loq note to AGENTS.md
* Deprecate add_tool_transformation and tool_transformations param
* Address PR review feedback: remove redundant imports, fix path reference
* Add missing imports to code examples in v3-features.mdx
* Add poll_interval to TaskConfig
Allow users to configure polling interval per component via
TaskConfig(poll_interval=timedelta(...)). Default is 5 seconds.
* Update snapshots for poll_interval field
* Add version badge to poll_interval docs
* Add defensive handling for Redis data and align default poll intervals
* Add AnthropicSamplingHandler
Adds a sampling handler for the Anthropic API at
fastmcp.client.sampling.handlers.anthropic, alongside the existing
OpenAI handler. Includes full support for tool calling.
Install with: pip install fastmcp[anthropic]
* Update default model
* Update sampling docs to cover both OpenAI and Anthropic handlers
* Use AsyncAnthropic, fix falsy value handling, handle tool_choice none
* Propagate isError to Anthropic, join multiple text blocks, fix docs
* Unify SamplingHandler and promote OpenAI handler
Consolidates ServerSamplingHandler and ClientSamplingHandler into a single
SamplingHandler type alias. Moves OpenAISamplingHandler from experimental
to fastmcp.client.sampling.handlers.openai as the canonical location.
Backwards compatibility maintained for imports from experimental.
* Remove unreachable code paths in OpenAI handler
* Fix docstring and use elif for mutually exclusive branches
* MCP → SDK (vocab change only)
* WIP: Sampling API with SamplingResult[T] and result_type
* SEP-1577: Sampling with tools
- Add tools and result_type parameters to ctx.sample()
- Update OpenAI handler for tool content types
- Client advertises sampling.tools capability by default
- Collect tool results into single message with list content
* Fix tool result content handling in OpenAI handler
* Remove @sampling_tool decorator - pass functions directly to sample()
Functions passed to ctx.sample(tools=[...]) are now auto-converted
via SamplingTool.from_function(). Users can still use that method
directly for custom name/description overrides.
* Remove auto-conversion of MCP tools to sampling tools
Users want MCP tools passed to ctx.sample() to go through the full MCP
machinery (middleware, native responses) rather than being auto-converted
to direct function calls. Now only SamplingTool and plain callables are
accepted - passing a FastMCP Tool raises a clear TypeError.
Also bumps mcp dependency to >=1.24.0 for required sampling features.
* Refactor sampling API: replace sample_iter() with sample_step()
Replace the mutable SampleRun/sample_iter() pattern with a simpler stateless
sample_step() function. sample_step() makes a single LLM call and returns a
SampleStep with the response and history. sample() now loops sample_step()
internally.
Key changes:
- Add sample_step() for fine-grained control over the sampling loop
- Remove SampleRun class and sample_iter() method
- Structured output uses tool description only (no prompt modification)
- execute_tools parameter controls automatic vs manual tool execution
* Address CodeRabbit nitpicks
* Address CodeRabbit review feedback for sampling tools
- Fix temperature=0.0 being dropped due to falsy evaluation
- Add ToolChoice.name support for forcing specific tools
- Replace assert statements with explicit RuntimeError checks
- Add mask_error_details parameter to sample()/sample_step() with ToolError escape hatch
- Fix hasattr patterns with proper isinstance checks
- Document mask_error_details and add OpenAI prerequisites to docs
* Address additional CodeRabbit review feedback
- Catch ValidationError specifically instead of bare Exception
- Update result_type docs to mention dataclasses and basic types
- Raise ValueError for unknown tool_choice modes
- Validate sampling_handler_behavior to catch typos
- Remove ToolChoice.name handling (not part of MCP spec)
- Validate tool_choice string in sample_step()
* Review fixes for sampling tools PR
- Remove internal functions from sampling __init__.py exports
- Remove fragile is_text property, use not is_tool_use instead
- Inline call_client into context.py, remove from run.py
- Fix SamplingMessage docs to use TextContent
- Handle result.text being None in doc examples
- Simplify client sampling docs to recommend OpenAISamplingHandler
- Add sampling_capabilities override documentation
- Raise iteration limit from 50 to 100
- Remove _parse_model_preferences duplication
- Use AsyncOpenAI in OpenAISamplingHandler
- Fix tool_choice docstring
* Fix OpenAI handler tests to use AsyncOpenAI
* Address remaining CodeRabbit review comments
- Fix message ordering in OpenAI handler: tool results now correctly
follow assistant message with tool_calls
- sample_step() now always includes assistant message in history
- Raise ValueError on JSON parse errors instead of silent {}
- Add has_sampling capability check when behavior is None
- Raise RuntimeError when structured output receives text response
- Wrap primitive result_type schemas in object wrapper
- Fix docs example using invalid SamplingMessage construction
- Add comprehensive client_sampling_test.py example
* Add return type annotation to OpenAISamplingHandler.__init__
* Use explicit 'is not None' check for sampling_capabilities defaulting
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Bill Easton <strawgate@users.noreply.github.com>
