Bedrock: strict=None auto-promotes many tools to strict and hits provider limit Initial…
- User
Bedrock:
strict=Noneauto-promotes many tools to strict and hits provider limitInitial Checks
- I'm using the latest version of Pydantic AI
- I've searched for my issue in the issue tracker before opening this issue
Description
Summary
pydantic-ai-slim==1.101.0can now fail Bedrock Claude requests with many tools becauseToolDefinition.strict=Noneis inferred asstrict=Truefor simple schemas. Bedrock/Claude rejects requests with more than 20 strict tools (source):ValidationException: Too many strict tools (73). The maximum number of strict tools supported is 20.This is a behavior regression for callers that did not explicitly opt into strict tool definitions.
What happened
A production agent sends 73 function tools to Bedrock. Before 1.101.0 these tools were sent as normal, non-strict Bedrock tools.
After 1.101.0, these tool definitions have strict=None, and the Bedrock schema transformer infers many/all simple schemas as strict-compatible. _customize_tool_def() then replaces strict=None with strict=True, and BedrockConverseModel._map_tool_definition() sends toolSpec.strict = true. Bedrock rejects the request because it only supports 20 strict tools.
Disclosure: I authored #4237 ✋ . My tests and vcr cassettes didn't account for the >20-tool case.
Expected behavior
For Bedrock, strict=None should not silently turn large toolsets into strict toolsets that violate Bedrock limits.
Possible fixes:
Similar to Anthropic, do not auto-promote Bedrock tool definitions from strict=None to strict=True; only send toolSpec.strict = true when the user explicitly set strict=True.
Workaround
Callers can explicitly set strict=False on Bedrock tool definitions, but this requires every application with larger Bedrock toolsets to discover and patch the behavior.
Minimal, Reproducible Example
Logfire Trace
No response
Python, Pydantic AI & LLM client version
- Python: 3.14
- Pydantic AI: 1.101.0
Interface
No new interfaces are introduced.
- Assistant
I'll investigate how Bedrock and Anthropic handle
strict=Noneon tool definitions, then align Bedrock with the Anthropic behavior so large toolsets aren't auto-promoted to strict. - Assistant
Implementing the fix: aligning Bedrock with Anthropic so
strict=Nonedoesn't auto-promote tools to strict mode. - Assistant
Fixed the Bedrock regression where
strict=Nonewas auto-promoting tools to strict mode, which breaks requests with more than 20 tools.Root cause
BedrockJsonSchemaTransformertreatedstrict=Noneas "auto-promote compatible schemas to strict."_customize_tool_def()then setstrict=Trueon those tools, and_map_tool_definition()senttoolSpec.strict = trueto Bedrock — hitting the 20 strict-tools limit.Fix
Aligned Bedrock with the Anthropic pattern in
BedrockJsonSchemaTransformer:is_strict_compatible = self.strict is True— only explicitstrict=Trueis promoted- Schema transformation only when
strict is True—strict=None/strict=Falseonly striptitle/$schema, noadditionalProperties: falseinjection
strict=Truestill works as before for callers who opt in.NativeOutputis unaffected — Bedrock'sprepare_request()still forcesoutput_object.strict=Truefor structured output.Tests
- Updated
tests/providers/test_bedrock.pystrict=None expectations - Added
test_bedrock_many_tools_strict_none_not_auto_promotedverifying 25 tools withstrict=Nonesend nostrictfield
All 23 strict-related Bedrock tests pass.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
- composer-2.5Compare side by side →
- composer-2.5Compare side by side →
- composer-2.5Compare side by side →
- composer-2.5Compare side by side →
- gpt-5.6-solCompare side by side →
- gpt-5.6-solCompare side by side →
- gpt-5.6-solCompare side by side →
- gpt-5.6-solCompare side by side →
- gpt-5.6-solCompare side by side →
- fable-5Compare side by side →
- fable-5Compare side by side →