added

This change introduces the _loop_response_model method in BaseAgentExecutor to conditionally return the response_model based on the presence of tools. The CrewAgentExecutor and AgentExecutor classes have been updated to utilize this new method, ensuring that the response_model is not sent to the LLM during tool loops, which prevents issues with structured-output schemas. Additionally, tests have been added to verify the correct behavior of this implementation.

conftest.py

@@ -187,6 +187,8 @@ HEADERS_TO_FILTER = {
     "anthropic-ratelimit-tokens-remaining": "ANTHROPIC-RATELIMIT-TOKENS-REMAINING-XXX",
     "anthropic-ratelimit-tokens-reset": "ANTHROPIC-RATELIMIT-TOKENS-RESET-XXX",
     "x-amz-date": "X-AMZ-DATE-XXX",
+    "x-amz-security-token": "X-AMZ-SECURITY-TOKEN-XXX",
+    "x-crewai-organization-id": "X-CREWAI-ORGANIZATION-ID-XXX",
     "amz-sdk-invocation-id": "AMZ-SDK-INVOCATION-ID-XXX",
     "accept-encoding": "ACCEPT-ENCODING-XXX",
     "x-amzn-requestid": "X-AMZN-REQUESTID-XXX",

lib/crewai/src/crewai/agents/agent_builder/base_agent_executor.py

@@ -2,7 +2,7 @@ from __future__ import annotations
 
 from typing import TYPE_CHECKING
 
-from pydantic import BaseModel, Field, PrivateAttr
+from pydantic import BaseModel as PydanticBaseModel, Field, PrivateAttr
 
 from crewai.agents.parser import AgentFinish
 from crewai.memory.utils import sanitize_scope_name
@@ -16,7 +16,7 @@ if TYPE_CHECKING:
     from crewai.task import Task
 
 
-class BaseAgentExecutor(BaseModel):
+class BaseAgentExecutor(PydanticBaseModel):
     model_config = {"arbitrary_types_allowed": True}
 
     executor_type: str = "base"
@@ -28,6 +28,26 @@ class BaseAgentExecutor(BaseModel):
     messages: list[LLMMessage] = Field(default_factory=list)
     _resuming: bool = PrivateAttr(default=False)
 
+    def _loop_response_model(self) -> type[PydanticBaseModel] | None:
+        """Return the response_model to forward to the LLM during a tool loop.
+
+        When the executor has tools, returns ``None``. Sending a strict
+        structured-output schema alongside ``tools=`` makes many providers
+        (notably OpenAI's Responses API with ``text.format`` json_schema)
+        constrain the first response to the schema, so the model emits
+        placeholder JSON instead of calling tools. Schema conversion is
+        applied as post-processing in ``Task._export_output()`` (Crew path)
+        or via ``Converter`` in ``Agent._build_output_from_result()``
+        (standalone kickoff path).
+
+        Subclasses must define ``original_tools`` and ``response_model``
+        attributes; this base implementation reads them via ``getattr`` so
+        the helper works for any executor that exposes those names.
+        """
+        if getattr(self, "original_tools", None):
+            return None
+        return getattr(self, "response_model", None)
+
     def _save_to_memory(self, output: AgentFinish) -> None:
         """Save task result to unified memory (memory or crew._memory)."""
         if self.agent is None:

lib/crewai/src/crewai/agents/crew_agent_executor.py

@@ -334,6 +334,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
 
                 enforce_rpm_limit(self.request_within_rpm_limit)
 
+                loop_response_model = self._loop_response_model()
                 answer = get_llm_response(
                     llm=cast("BaseLLM", self.llm),
                     messages=self.messages,
@@ -341,11 +342,11 @@ class CrewAgentExecutor(BaseAgentExecutor):
                     printer=PRINTER,
                     from_task=self.task,
                     from_agent=self.agent,
-                    response_model=self.response_model,
+                    response_model=loop_response_model,
                     executor_context=self,
                     verbose=self.agent.verbose,
                 )
-                if self.response_model is not None:
+                if loop_response_model is not None:
                     try:
                         if isinstance(answer, BaseModel):
                             output_json = answer.model_dump_json()
@@ -355,7 +356,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                                 text=output_json,
                             )
                         else:
-                            self.response_model.model_validate_json(answer)
+                            loop_response_model.model_validate_json(answer)
                             formatted_answer = AgentFinish(
                                 thought="",
                                 output=answer,
@@ -486,7 +487,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                     available_functions=None,
                     from_task=self.task,
                     from_agent=self.agent,
-                    response_model=self.response_model,
+                    response_model=self._loop_response_model(),
                     executor_context=self,
                     verbose=self.agent.verbose,
                 )
@@ -1142,6 +1143,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
 
                 enforce_rpm_limit(self.request_within_rpm_limit)
 
+                loop_response_model = self._loop_response_model()
                 answer = await aget_llm_response(
                     llm=cast("BaseLLM", self.llm),
                     messages=self.messages,
@@ -1149,12 +1151,12 @@ class CrewAgentExecutor(BaseAgentExecutor):
                     printer=PRINTER,
                     from_task=self.task,
                     from_agent=self.agent,
-                    response_model=self.response_model,
+                    response_model=loop_response_model,
                     executor_context=self,
                     verbose=self.agent.verbose,
                 )
 
-                if self.response_model is not None:
+                if loop_response_model is not None:
                     try:
                         if isinstance(answer, BaseModel):
                             output_json = answer.model_dump_json()
@@ -1164,7 +1166,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                                 text=output_json,
                             )
                         else:
-                            self.response_model.model_validate_json(answer)
+                            loop_response_model.model_validate_json(answer)
                             formatted_answer = AgentFinish(
                                 thought="",
                                 output=answer,
@@ -1295,7 +1297,7 @@ class CrewAgentExecutor(BaseAgentExecutor):
                     available_functions=None,
                     from_task=self.task,
                     from_agent=self.agent,
-                    response_model=self.response_model,
+                    response_model=self._loop_response_model(),
                     executor_context=self,
                     verbose=self.agent.verbose,
                 )

lib/crewai/src/crewai/experimental/agent_executor.py

@@ -1229,7 +1229,7 @@ class AgentExecutor(Flow[AgentExecutorState], BaseAgentExecutor):  # type: ignor
                 printer=PRINTER,
                 from_task=self.task,
                 from_agent=self.agent,
-                response_model=self.response_model,
+                response_model=self._loop_response_model(),
                 executor_context=self,
                 verbose=self.agent.verbose,
             )
@@ -1317,7 +1317,7 @@ class AgentExecutor(Flow[AgentExecutorState], BaseAgentExecutor):  # type: ignor
                 available_functions=None,
                 from_task=self.task,
                 from_agent=self.agent,
-                response_model=self.response_model,
+                response_model=self._loop_response_model(),
                 executor_context=self,
                 verbose=self.agent.verbose,
             )

lib/crewai/tests/agents/test_agent_executor.py

@@ -2044,3 +2044,103 @@ class TestVisionImageFormatContract:
         assert hasattr(AnthropicCompletion, "_convert_image_blocks"), (
             "Anthropic provider must have _convert_image_blocks for auto-conversion"
         )
+
+
+class TestLoopResponseModelGuard:
+    """Regression: response_model must not be sent to the LLM during a tool loop.
+
+    Bug: when output_pydantic + tools are both set, OpenAI's Responses API
+    (and several other providers) constrain the first response to the
+    structured-output schema, so the model emits placeholder JSON instead of
+    calling tools. Schema conversion is post-processed, so the LLM call must
+    omit response_model whenever tools are present.
+    """
+
+    def _make_pydantic_model(self):
+        from pydantic import BaseModel as _BaseModel
+
+        class _Out(_BaseModel):
+            answer: str
+
+        return _Out
+
+    def _crew_executor(self, *, with_tools: bool, response_model):
+        from crewai.agents.crew_agent_executor import CrewAgentExecutor
+
+        tool = Mock()
+        tool.name = "search"
+        tool.description = "search"
+
+        executor = CrewAgentExecutor.model_construct(
+            response_model=response_model,
+            original_tools=[tool] if with_tools else [],
+        )
+        return executor
+
+    def test_crew_helper_strips_when_tools_present(self):
+        Model = self._make_pydantic_model()
+        ex = self._crew_executor(with_tools=True, response_model=Model)
+        assert ex._loop_response_model() is None
+
+    def test_crew_helper_preserves_when_no_tools(self):
+        Model = self._make_pydantic_model()
+        ex = self._crew_executor(with_tools=False, response_model=Model)
+        assert ex._loop_response_model() is Model
+
+    def test_crew_helper_returns_none_when_no_response_model(self):
+        ex = self._crew_executor(with_tools=False, response_model=None)
+        assert ex._loop_response_model() is None
+
+    def test_experimental_helper_strips_when_tools_present(self):
+        Model = self._make_pydantic_model()
+        llm = Mock()
+        llm.supports_stop_words.return_value = False
+        tool = Mock()
+        tool.name = "search"
+        ex = _build_executor(llm=llm, response_model=Model, original_tools=[tool])
+        assert ex._loop_response_model() is None
+
+    def test_experimental_helper_preserves_when_no_tools(self):
+        Model = self._make_pydantic_model()
+        llm = Mock()
+        llm.supports_stop_words.return_value = False
+        ex = _build_executor(llm=llm, response_model=Model, original_tools=[])
+        assert ex._loop_response_model() is Model
+
+    def test_loop_call_sites_route_through_helper(self):
+        """Source-level guard: every in-loop LLM call must use the helper.
+
+        This catches regressions where someone reintroduces
+        `response_model=self.response_model` inside a tool loop.
+        """
+        import inspect
+        from crewai.agents import crew_agent_executor as _crew_mod
+        from crewai.experimental import agent_executor as _exp_mod
+
+        for mod, names in (
+            (
+                _crew_mod,
+                (
+                    "_invoke_loop_react",
+                    "_invoke_loop_native_tools",
+                    "_ainvoke_loop_react",
+                    "_ainvoke_loop_native_tools",
+                ),
+            ),
+            (
+                _exp_mod,
+                ("call_llm_and_parse", "call_llm_native_tools"),
+            ),
+        ):
+            cls = (
+                _crew_mod.CrewAgentExecutor
+                if mod is _crew_mod
+                else _exp_mod.AgentExecutor
+            )
+            for name in names:
+                src = inspect.getsource(getattr(cls, name))
+                assert "response_model=self.response_model" not in src, (
+                    f"{cls.__name__}.{name} forwards self.response_model directly. "
+                    "Use self._loop_response_model() instead — see "
+                    "TestLoopResponseModelGuard for the bug it prevents."
+                )

lib/crewai/tests/cassettes/llms/anthropic/test_anthropic_agent_kickoff_structured_output_with_tools.yaml

@@ -1,14 +1,10 @@
 interactions:
 - request:
     body: '{"max_tokens":4096,"messages":[{"role":"user","content":"\nCurrent Task:
-      Calculate 15 + 27 using your add_numbers tool. Report the result."}],"model":"claude-3-5-haiku-20241022","stop_sequences":["\nObservation:"],"stream":false,"system":"You
+      Calculate 15 + 27 using your add_numbers tool. Report the result."}],"model":"claude-sonnet-4-6","stop_sequences":["\nObservation:"],"stream":false,"system":"You
       are Calculator. You are a calculator assistant that uses tools to compute results.\nYour
-      personal goal is: Perform calculations using available tools","tool_choice":{"type":"tool","name":"structured_output"},"tools":[{"name":"structured_output","description":"Output
-      the structured response","input_schema":{"type":"object","description":"Structured
-      output for calculation results.","title":"CalculationResult","properties":{"operation":{"type":"string","description":"The
-      mathematical operation performed","title":"Operation"},"result":{"type":"integer","description":"The
-      result of the calculation","title":"Result"},"explanation":{"type":"string","description":"Brief
-      explanation of the calculation","title":"Explanation"}},"additionalProperties":false,"required":["operation","result","explanation"]}}]}'
+      personal goal is: Perform calculations using available tools","tools":[{"name":"add_numbers","description":"Add
+      two numbers together and return the sum.","input_schema":{"properties":{"a":{"type":"integer"},"b":{"type":"integer"}},"required":["a","b"],"type":"object","additionalProperties":false},"strict":true}]}'
     headers:
       User-Agent:
       - X-USER-AGENT-XXX
@@ -21,7 +17,7 @@ interactions:
       connection:
       - keep-alive
       content-length:
-      - '1050'
+      - '638'
       content-type:
       - application/json
       host:
@@ -37,7 +33,7 @@ interactions:
       x-stainless-os:
       - X-STAINLESS-OS-XXX
       x-stainless-package-version:
-      - 0.73.0
+      - 0.96.0
       x-stainless-retry-count:
       - '0'
       x-stainless-runtime:
@@ -50,8 +46,8 @@ interactions:
     uri: https://api.anthropic.com/v1/messages
   response:
     body:
-      string: '{"model":"claude-3-5-haiku-20241022","id":"msg_01MsoNSVoPuoMYGCcJLvfXS6","type":"message","role":"assistant","content":[{"type":"tool_use","id":"toolu_01TtAsqddWjE7C4GYmCKavdg","name":"structured_output","input":{"operation":"Addition","result":42,"explanation":"Added
-        15 and 27 together"}}],"stop_reason":"tool_use","stop_sequence":null,"usage":{"input_tokens":573,"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"cache_creation":{"ephemeral_5m_input_tokens":0,"ephemeral_1h_input_tokens":0},"output_tokens":75,"service_tier":"standard","inference_geo":"not_available"}}'
+      string: '{"model":"claude-sonnet-4-6","id":"msg_016CHZY3K6G73PFc4iAJVC5C","type":"message","role":"assistant","content":[{"type":"text","text":"I''ll
+        calculate 15 + 27 right away using the `add_numbers` tool!"},{"type":"tool_use","id":"toolu_011UPTzkFNsgYshddoTX2A8X","name":"add_numbers","input":{"a":15,"b":27},"caller":{"type":"direct"}}],"stop_reason":"tool_use","stop_sequence":null,"stop_details":null,"usage":{"input_tokens":633,"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"cache_creation":{"ephemeral_5m_input_tokens":0,"ephemeral_1h_input_tokens":0},"output_tokens":92,"service_tier":"standard","inference_geo":"global"}}'
     headers:
       CF-RAY:
       - CF-RAY-XXX
@@ -62,7 +58,7 @@ interactions:
       Content-Type:
       - application/json
       Date:
-      - Thu, 12 Feb 2026 22:11:20 GMT
+      - Fri, 24 Apr 2026 20:48:28 GMT
       Server:
       - cloudflare
       Transfer-Encoding:
@@ -84,11 +80,119 @@ interactions:
       anthropic-ratelimit-output-tokens-reset:
       - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-RESET-XXX
       anthropic-ratelimit-requests-limit:
-      - '4000'
+      - '20000'
       anthropic-ratelimit-requests-remaining:
-      - '3999'
+      - '19999'
       anthropic-ratelimit-requests-reset:
-      - '2026-02-12T22:11:18Z'
+      - '2026-04-24T20:48:25Z'
+      anthropic-ratelimit-tokens-limit:
+      - ANTHROPIC-RATELIMIT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-tokens-reset:
+      - ANTHROPIC-RATELIMIT-TOKENS-RESET-XXX
+      cf-cache-status:
+      - DYNAMIC
+      request-id:
+      - REQUEST-ID-XXX
+      set-cookie:
+      - SET-COOKIE-XXX
+      strict-transport-security:
+      - STS-XXX
+      x-envoy-upstream-service-time:
+      - '2056'
+    status:
+      code: 200
+      message: OK
+- request:
+    body: '{"max_tokens":4096,"messages":[{"role":"user","content":"\nCurrent Task:
+      Calculate 15 + 27 using your add_numbers tool. Report the result."},{"role":"assistant","content":[{"type":"tool_use","id":"toolu_011UPTzkFNsgYshddoTX2A8X","name":"add_numbers","input":{"a":15,"b":27}}]},{"role":"user","content":[{"type":"tool_result","tool_use_id":"toolu_011UPTzkFNsgYshddoTX2A8X","content":"42"}]}],"model":"claude-sonnet-4-6","stop_sequences":["\nObservation:"],"stream":false,"system":"You
+      are Calculator. You are a calculator assistant that uses tools to compute results.\nYour
+      personal goal is: Perform calculations using available tools","tools":[{"name":"add_numbers","description":"Add
+      two numbers together and return the sum.","input_schema":{"properties":{"a":{"type":"integer"},"b":{"type":"integer"}},"required":["a","b"],"type":"object","additionalProperties":false},"strict":true}]}'
+    headers:
+      User-Agent:
+      - X-USER-AGENT-XXX
+      accept:
+      - application/json
+      accept-encoding:
+      - ACCEPT-ENCODING-XXX
+      anthropic-version:
+      - '2023-06-01'
+      connection:
+      - keep-alive
+      content-length:
+      - '887'
+      content-type:
+      - application/json
+      cookie:
+      - COOKIE-XXX
+      host:
+      - api.anthropic.com
+      x-api-key:
+      - X-API-KEY-XXX
+      x-stainless-arch:
+      - X-STAINLESS-ARCH-XXX
+      x-stainless-async:
+      - 'false'
+      x-stainless-lang:
+      - python
+      x-stainless-os:
+      - X-STAINLESS-OS-XXX
+      x-stainless-package-version:
+      - 0.96.0
+      x-stainless-retry-count:
+      - '0'
+      x-stainless-runtime:
+      - CPython
+      x-stainless-runtime-version:
+      - 3.13.3
+      x-stainless-timeout:
+      - NOT_GIVEN
+    method: POST
+    uri: https://api.anthropic.com/v1/messages
+  response:
+    body:
+      string: "{\"model\":\"claude-sonnet-4-6\",\"id\":\"msg_01V13FHBygrTHV6pnhocqvd9\",\"type\":\"message\",\"role\":\"assistant\",\"content\":[{\"type\":\"text\",\"text\":\"The
+        result of **15 + 27 = 42**! \U0001F389\\n\\nUsing the `add_numbers` tool,
+        I computed the sum of 15 and 27, and the result is **42**.\"}],\"stop_reason\":\"end_turn\",\"stop_sequence\":null,\"stop_details\":null,\"usage\":{\"input_tokens\":717,\"cache_creation_input_tokens\":0,\"cache_read_input_tokens\":0,\"cache_creation\":{\"ephemeral_5m_input_tokens\":0,\"ephemeral_1h_input_tokens\":0},\"output_tokens\":51,\"service_tier\":\"standard\",\"inference_geo\":\"global\"}}"
+    headers:
+      CF-RAY:
+      - CF-RAY-XXX
+      Connection:
+      - keep-alive
+      Content-Security-Policy:
+      - CSP-FILTERED
+      Content-Type:
+      - application/json
+      Date:
+      - Fri, 24 Apr 2026 20:48:32 GMT
+      Server:
+      - cloudflare
+      Transfer-Encoding:
+      - chunked
+      X-Robots-Tag:
+      - none
+      anthropic-organization-id:
+      - ANTHROPIC-ORGANIZATION-ID-XXX
+      anthropic-ratelimit-input-tokens-limit:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-input-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-input-tokens-reset:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-RESET-XXX
+      anthropic-ratelimit-output-tokens-limit:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-output-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-output-tokens-reset:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-RESET-XXX
+      anthropic-ratelimit-requests-limit:
+      - '20000'
+      anthropic-ratelimit-requests-remaining:
+      - '19999'
+      anthropic-ratelimit-requests-reset:
+      - '2026-04-24T20:48:28Z'
       anthropic-ratelimit-tokens-limit:
       - ANTHROPIC-RATELIMIT-TOKENS-LIMIT-XXX
       anthropic-ratelimit-tokens-remaining:
@@ -102,7 +206,139 @@ interactions:
       strict-transport-security:
       - STS-XXX
       x-envoy-upstream-service-time:
-      - '1234'
+      - '4700'
+    status:
+      code: 200
+      message: OK
+- request:
+    body: "{\"max_tokens\":4096,\"messages\":[{\"role\":\"user\",\"content\":\"The
+      result of **15 + 27 = 42**! \U0001F389\\n\\nUsing the `add_numbers` tool, I
+      computed the sum of 15 and 27, and the result is **42**.\"}],\"model\":\"claude-sonnet-4-6\",\"stop_sequences\":[\"\\nObservation:\"],\"stream\":false,\"system\":\"Format
+      your final answer according to the following OpenAPI schema: {\\n  \\\"type\\\":
+      \\\"json_schema\\\",\\n  \\\"json_schema\\\": {\\n    \\\"name\\\": \\\"CalculationResult\\\",\\n
+      \   \\\"strict\\\": true,\\n    \\\"schema\\\": {\\n      \\\"description\\\":
+      \\\"Structured output for calculation results.\\\",\\n      \\\"properties\\\":
+      {\\n        \\\"operation\\\": {\\n          \\\"description\\\": \\\"The mathematical
+      operation performed\\\",\\n          \\\"title\\\": \\\"Operation\\\",\\n          \\\"type\\\":
+      \\\"string\\\"\\n        },\\n        \\\"result\\\": {\\n          \\\"description\\\":
+      \\\"The result of the calculation\\\",\\n          \\\"title\\\": \\\"Result\\\",\\n
+      \         \\\"type\\\": \\\"integer\\\"\\n        },\\n        \\\"explanation\\\":
+      {\\n          \\\"description\\\": \\\"Brief explanation of the calculation\\\",\\n
+      \         \\\"title\\\": \\\"Explanation\\\",\\n          \\\"type\\\": \\\"string\\\"\\n
+      \       }\\n      },\\n      \\\"required\\\": [\\n        \\\"operation\\\",\\n
+      \       \\\"result\\\",\\n        \\\"explanation\\\"\\n      ],\\n      \\\"title\\\":
+      \\\"CalculationResult\\\",\\n      \\\"type\\\": \\\"object\\\",\\n      \\\"additionalProperties\\\":
+      false\\n    }\\n  }\\n}\\n\\nIMPORTANT: Preserve the original content exactly
+      as-is. Do NOT rewrite, paraphrase, or modify the meaning of the content. Only
+      structure it to match the schema format.\\n\\nDo not include the OpenAPI schema
+      in the final output. Ensure the final output does not include any code block
+      markers like ```json or ```python.\",\"tool_choice\":{\"type\":\"tool\",\"name\":\"structured_output\"},\"tools\":[{\"name\":\"structured_output\",\"description\":\"Output
+      the structured response\",\"input_schema\":{\"type\":\"object\",\"description\":\"Structured
+      output for calculation results.\",\"title\":\"CalculationResult\",\"properties\":{\"operation\":{\"type\":\"string\",\"description\":\"The
+      mathematical operation performed\",\"title\":\"Operation\"},\"result\":{\"type\":\"integer\",\"description\":\"The
+      result of the calculation\",\"title\":\"Result\"},\"explanation\":{\"type\":\"string\",\"description\":\"Brief
+      explanation of the calculation\",\"title\":\"Explanation\"}},\"additionalProperties\":false,\"required\":[\"operation\",\"result\",\"explanation\"]}}]}"
+    headers:
+      User-Agent:
+      - X-USER-AGENT-XXX
+      accept:
+      - application/json
+      accept-encoding:
+      - ACCEPT-ENCODING-XXX
+      anthropic-version:
+      - '2023-06-01'
+      connection:
+      - keep-alive
+      content-length:
+      - '2313'
+      content-type:
+      - application/json
+      cookie:
+      - COOKIE-XXX
+      host:
+      - api.anthropic.com
+      x-api-key:
+      - X-API-KEY-XXX
+      x-stainless-arch:
+      - X-STAINLESS-ARCH-XXX
+      x-stainless-async:
+      - 'false'
+      x-stainless-lang:
+      - python
+      x-stainless-os:
+      - X-STAINLESS-OS-XXX
+      x-stainless-package-version:
+      - 0.96.0
+      x-stainless-retry-count:
+      - '0'
+      x-stainless-runtime:
+      - CPython
+      x-stainless-runtime-version:
+      - 3.13.3
+      x-stainless-timeout:
+      - NOT_GIVEN
+    method: POST
+    uri: https://api.anthropic.com/v1/messages
+  response:
+    body:
+      string: "{\"model\":\"claude-sonnet-4-6\",\"id\":\"msg_01Csx2k6XmGyjfWkdLrXzLyA\",\"type\":\"message\",\"role\":\"assistant\",\"content\":[{\"type\":\"tool_use\",\"id\":\"toolu_01DzJ3ZUUnvPPJgzioZnxjKM\",\"name\":\"structured_output\",\"input\":{\"operation\":\"add_numbers\",\"result\":42,\"explanation\":\"The
+        result of **15 + 27 = 42**! \U0001F389\\n\\nUsing the `add_numbers` tool,
+        I computed the sum of 15 and 27, and the result is **42**.\"},\"caller\":{\"type\":\"direct\"}}],\"stop_reason\":\"tool_use\",\"stop_sequence\":null,\"stop_details\":null,\"usage\":{\"input_tokens\":1122,\"cache_creation_input_tokens\":0,\"cache_read_input_tokens\":0,\"cache_creation\":{\"ephemeral_5m_input_tokens\":0,\"ephemeral_1h_input_tokens\":0},\"output_tokens\":116,\"service_tier\":\"standard\",\"inference_geo\":\"global\"}}"
+    headers:
+      CF-RAY:
+      - CF-RAY-XXX
+      Connection:
+      - keep-alive
+      Content-Security-Policy:
+      - CSP-FILTERED
+      Content-Type:
+      - application/json
+      Date:
+      - Fri, 24 Apr 2026 20:48:34 GMT
+      Server:
+      - cloudflare
+      Transfer-Encoding:
+      - chunked
+      X-Robots-Tag:
+      - none
+      anthropic-organization-id:
+      - ANTHROPIC-ORGANIZATION-ID-XXX
+      anthropic-ratelimit-input-tokens-limit:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-input-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-input-tokens-reset:
+      - ANTHROPIC-RATELIMIT-INPUT-TOKENS-RESET-XXX
+      anthropic-ratelimit-output-tokens-limit:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-output-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-output-tokens-reset:
+      - ANTHROPIC-RATELIMIT-OUTPUT-TOKENS-RESET-XXX
+      anthropic-ratelimit-requests-limit:
+      - '20000'
+      anthropic-ratelimit-requests-remaining:
+      - '19999'
+      anthropic-ratelimit-requests-reset:
+      - '2026-04-24T20:48:32Z'
+      anthropic-ratelimit-tokens-limit:
+      - ANTHROPIC-RATELIMIT-TOKENS-LIMIT-XXX
+      anthropic-ratelimit-tokens-remaining:
+      - ANTHROPIC-RATELIMIT-TOKENS-REMAINING-XXX
+      anthropic-ratelimit-tokens-reset:
+      - ANTHROPIC-RATELIMIT-TOKENS-RESET-XXX
+      cf-cache-status:
+      - DYNAMIC
+      request-id:
+      - REQUEST-ID-XXX
+      server-timing:
+      - x-originResponse;dur=1594
+      strict-transport-security:
+      - STS-XXX
+      vary:
+      - Accept-Encoding
+      x-envoy-upstream-service-time:
+      - '1588'
     status:
       code: 200
       message: OK

lib/crewai/tests/cassettes/llms/bedrock/test_bedrock_agent_kickoff_structured_output_with_tools.yaml

@@ -8,20 +8,10 @@ interactions:
       [{"toolSpec": {"name": "add_numbers", "description": "Add two numbers together
       and return the sum.", "inputSchema": {"json": {"properties": {"a": {"title":
       "A", "type": "integer"}, "b": {"title": "B", "type": "integer"}}, "required":
-      ["a", "b"], "type": "object", "additionalProperties": false}}}}, {"toolSpec":
-      {"name": "structured_output", "description": "Use this tool to provide your
-      final structured response. Call this tool when you have gathered all necessary
-      information and are ready to provide the final answer in the required format.",
-      "inputSchema": {"json": {"description": "Structured output for calculation results.",
-      "properties": {"operation": {"description": "The mathematical operation performed",
-      "title": "Operation", "type": "string"}, "result": {"description": "The result
-      of the calculation", "title": "Result", "type": "integer"}, "explanation": {"description":
-      "Brief explanation of the calculation", "title": "Explanation", "type": "string"}},
-      "required": ["operation", "result", "explanation"], "title": "CalculationResult",
-      "type": "object", "additionalProperties": false}}}}]}}'
+      ["a", "b"], "type": "object", "additionalProperties": false}}}}]}}'
     headers:
       Content-Length:
-      - '1509'
+      - '702'
       Content-Type:
       - !!binary |
         YXBwbGljYXRpb24vanNvbg==
@@ -36,21 +26,23 @@ interactions:
       - AUTHORIZATION-XXX
       x-amz-date:
       - X-AMZ-DATE-XXX
+      x-amz-security-token:
+      - X-AMZ-SECURITY-TOKEN-XXX
     method: POST
-    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/anthropic.claude-3-sonnet-20240229-v1%3A0/converse
+    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/us.anthropic.claude-sonnet-4-6/converse
   response:
     body:
-      string: '{"metrics":{"latencyMs":1968},"output":{"message":{"content":[{"text":"Okay,
-        let''s calculate 15 + 27 using the add_numbers tool:"},{"toolUse":{"input":{"a":15,"b":27},"name":"add_numbers","toolUseId":"tooluse_pSseOamVELzpL3kQG5VukN"}}],"role":"assistant"}},"stopReason":"tool_use","usage":{"inputTokens":488,"outputTokens":91,"serverToolUsage":{},"totalTokens":579}}'
+      string: '{"metrics":{"latencyMs":1570},"output":{"message":{"content":[{"text":"I''ll
+        calculate 15 + 27 right away using the `add_numbers` tool!"},{"toolUse":{"input":{"a":15,"b":27},"name":"add_numbers","toolUseId":"tooluse_lYs0ulbVGQaYseRBvZ7tdX","type":"tool_use"}}],"role":"assistant"}},"stopReason":"tool_use","usage":{"cacheReadInputTokenCount":0,"cacheReadInputTokens":0,"cacheWriteInputTokenCount":0,"cacheWriteInputTokens":0,"inputTokens":645,"outputTokens":92,"serverToolUsage":{},"totalTokens":737}}'
     headers:
       Connection:
       - keep-alive
       Content-Length:
-      - '366'
+      - '500'
       Content-Type:
       - application/json
       Date:
-      - Thu, 12 Feb 2026 22:01:04 GMT
+      - Fri, 24 Apr 2026 20:49:36 GMT
       x-amzn-RequestId:
       - X-AMZN-REQUESTID-XXX
     status:
@@ -59,29 +51,19 @@ interactions:
 - request:
     body: '{"messages": [{"role": "user", "content": [{"text": "\nCurrent Task: Calculate
       15 + 27 using your add_numbers tool. Report the result."}]}, {"role": "assistant",
-      "content": [{"toolUse": {"toolUseId": "tooluse_pSseOamVELzpL3kQG5VukN", "name":
+      "content": [{"toolUse": {"toolUseId": "tooluse_lYs0ulbVGQaYseRBvZ7tdX", "name":
       "add_numbers", "input": {"a": 15, "b": 27}}}]}, {"role": "user", "content":
-      [{"toolResult": {"toolUseId": "tooluse_pSseOamVELzpL3kQG5VukN", "content": [{"text":
+      [{"toolResult": {"toolUseId": "tooluse_lYs0ulbVGQaYseRBvZ7tdX", "content": [{"text":
       "42"}]}}]}], "inferenceConfig": {"stopSequences": ["\nObservation:"]}, "system":
       [{"text": "You are Calculator. You are a calculator assistant that uses tools
       to compute results.\nYour personal goal is: Perform calculations using available
       tools"}], "toolConfig": {"tools": [{"toolSpec": {"name": "add_numbers", "description":
       "Add two numbers together and return the sum.", "inputSchema": {"json": {"properties":
       {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "integer"}},
-      "required": ["a", "b"], "type": "object", "additionalProperties": false}}}},
-      {"toolSpec": {"name": "structured_output", "description": "Use this tool to
-      provide your final structured response. Call this tool when you have gathered
-      all necessary information and are ready to provide the final answer in the required
-      format.", "inputSchema": {"json": {"description": "Structured output for calculation
-      results.", "properties": {"operation": {"description": "The mathematical operation
-      performed", "title": "Operation", "type": "string"}, "result": {"description":
-      "The result of the calculation", "title": "Result", "type": "integer"}, "explanation":
-      {"description": "Brief explanation of the calculation", "title": "Explanation",
-      "type": "string"}}, "required": ["operation", "result", "explanation"], "title":
-      "CalculationResult", "type": "object", "additionalProperties": false}}}}]}}'
+      "required": ["a", "b"], "type": "object", "additionalProperties": false}}}}]}}'
     headers:
       Content-Length:
-      - '1784'
+      - '977'
       Content-Type:
       - !!binary |
         YXBwbGljYXRpb24vanNvbg==
@@ -96,41 +78,50 @@ interactions:
       - AUTHORIZATION-XXX
       x-amz-date:
       - X-AMZ-DATE-XXX
+      x-amz-security-token:
+      - X-AMZ-SECURITY-TOKEN-XXX
     method: POST
-    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/anthropic.claude-3-sonnet-20240229-v1%3A0/converse
+    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/us.anthropic.claude-sonnet-4-6/converse
   response:
     body:
-      string: '{"metrics":{"latencyMs":7598},"output":{"message":{"content":[{"toolUse":{"input":{"operation":"Addition","result":42,"explanation":"I
-        added 15 and 27 using the add_numbers tool."},"name":"structured_output","toolUseId":"tooluse_RT8uSPaM37Q8CVuo3rJLtI"}}],"role":"assistant"}},"stopReason":"tool_use","usage":{"inputTokens":571,"outputTokens":102,"serverToolUsage":{},"totalTokens":673}}'
+      string: "{\"metrics\":{\"latencyMs\":1456},\"output\":{\"message\":{\"content\":[{\"text\":\"The
+        result of **15 + 27 = 42**! \U0001F9EE\\n\\nUsing the `add_numbers` tool,
+        I computed the sum of 15 and 27, and the result is **42**.\"}],\"role\":\"assistant\"}},\"stopReason\":\"end_turn\",\"usage\":{\"cacheReadInputTokenCount\":0,\"cacheReadInputTokens\":0,\"cacheWriteInputTokenCount\":0,\"cacheWriteInputTokens\":0,\"inputTokens\":729,\"outputTokens\":51,\"serverToolUsage\":{},\"totalTokens\":780}}"
     headers:
       Connection:
       - keep-alive
       Content-Length:
-      - '387'
+      - '443'
       Content-Type:
       - application/json
       Date:
-      - Thu, 12 Feb 2026 22:01:12 GMT
+      - Fri, 24 Apr 2026 20:49:37 GMT
       x-amzn-RequestId:
       - X-AMZN-REQUESTID-XXX
     status:
       code: 200
       message: OK
 - request:
-    body: '{"messages": [{"role": "user", "content": [{"text": "\nCurrent Task: Calculate
-      15 + 27 using your add_numbers tool. Report the result."}]}, {"role": "assistant",
-      "content": [{"toolUse": {"toolUseId": "tooluse_pSseOamVELzpL3kQG5VukN", "name":
-      "add_numbers", "input": {"a": 15, "b": 27}}}]}, {"role": "user", "content":
-      [{"toolResult": {"toolUseId": "tooluse_pSseOamVELzpL3kQG5VukN", "content": [{"text":
-      "42"}]}}]}, {"role": "assistant", "content": [{"text": "{\"operation\":\"Addition\",\"result\":42,\"explanation\":\"I
-      added 15 and 27 using the add_numbers tool.\"}"}]}], "inferenceConfig": {"stopSequences":
-      ["\nObservation:"]}, "system": [{"text": "You are Calculator. You are a calculator
-      assistant that uses tools to compute results.\nYour personal goal is: Perform
-      calculations using available tools"}], "toolConfig": {"tools": [{"toolSpec":
-      {"name": "add_numbers", "description": "Add two numbers together and return
-      the sum.", "inputSchema": {"json": {"properties": {"a": {"title": "A", "type":
-      "integer"}, "b": {"title": "B", "type": "integer"}}, "required": ["a", "b"],
-      "type": "object", "additionalProperties": false}}}}, {"toolSpec": {"name": "structured_output",
+    body: '{"messages": [{"role": "user", "content": [{"text": "The result of **15
+      + 27 = 42**! \ud83e\uddee\n\nUsing the `add_numbers` tool, I computed the sum
+      of 15 and 27, and the result is **42**."}]}], "inferenceConfig": {"stopSequences":
+      ["\nObservation:"]}, "system": [{"text": "Format your final answer according
+      to the following OpenAPI schema: {\n  \"type\": \"json_schema\",\n  \"json_schema\":
+      {\n    \"name\": \"CalculationResult\",\n    \"strict\": true,\n    \"schema\":
+      {\n      \"description\": \"Structured output for calculation results.\",\n      \"properties\":
+      {\n        \"operation\": {\n          \"description\": \"The mathematical operation
+      performed\",\n          \"title\": \"Operation\",\n          \"type\": \"string\"\n        },\n        \"result\":
+      {\n          \"description\": \"The result of the calculation\",\n          \"title\":
+      \"Result\",\n          \"type\": \"integer\"\n        },\n        \"explanation\":
+      {\n          \"description\": \"Brief explanation of the calculation\",\n          \"title\":
+      \"Explanation\",\n          \"type\": \"string\"\n        }\n      },\n      \"required\":
+      [\n        \"operation\",\n        \"result\",\n        \"explanation\"\n      ],\n      \"title\":
+      \"CalculationResult\",\n      \"type\": \"object\",\n      \"additionalProperties\":
+      false\n    }\n  }\n}\n\nIMPORTANT: Preserve the original content exactly as-is.
+      Do NOT rewrite, paraphrase, or modify the meaning of the content. Only structure
+      it to match the schema format.\n\nDo not include the OpenAPI schema in the final
+      output. Ensure the final output does not include any code block markers like
+      ```json or ```python."}], "toolConfig": {"tools": [{"toolSpec": {"name": "structured_output",
       "description": "Use this tool to provide your final structured response. Call
       this tool when you have gathered all necessary information and are ready to
       provide the final answer in the required format.", "inputSchema": {"json": {"description":
@@ -140,10 +131,10 @@ interactions:
       "type": "integer"}, "explanation": {"description": "Brief explanation of the
       calculation", "title": "Explanation", "type": "string"}}, "required": ["operation",
       "result", "explanation"], "title": "CalculationResult", "type": "object", "additionalProperties":
-      false}}}}]}}'
+      false}}}}], "toolChoice": {"tool": {"name": "structured_output"}}}}'
     headers:
       Content-Length:
-      - '1942'
+      - '2545'
       Content-Type:
       - !!binary |
         YXBwbGljYXRpb24vanNvbg==
@@ -158,28 +149,27 @@ interactions:
       - AUTHORIZATION-XXX
       x-amz-date:
       - X-AMZ-DATE-XXX
+      x-amz-security-token:
+      - X-AMZ-SECURITY-TOKEN-XXX
     method: POST
-    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/anthropic.claude-3-sonnet-20240229-v1%3A0/converse
+    uri: https://bedrock-runtime.us-east-1.amazonaws.com/model/us.anthropic.claude-sonnet-4-6/converse
   response:
     body:
-      string: '{"message":"The model returned the following errors: Your API request
-        included an `assistant` message in the final position, which would pre-fill
-        the `assistant` response. When using tools, pre-filling the `assistant` response
-        is not supported."}'
+      string: "{\"metrics\":{\"latencyMs\":1729},\"output\":{\"message\":{\"content\":[{\"toolUse\":{\"input\":{\"operation\":\"add_numbers\",\"result\":42,\"explanation\":\"The
+        result of **15 + 27 = 42**! \U0001F9EE\\n\\nUsing the `add_numbers` tool,
+        I computed the sum of 15 and 27, and the result is **42**.\"},\"name\":\"structured_output\",\"toolUseId\":\"tooluse_FavcTtQCbzFrqURMiduSla\",\"type\":\"tool_use\"}}],\"role\":\"assistant\"}},\"stopReason\":\"tool_use\",\"usage\":{\"cacheReadInputTokenCount\":0,\"cacheReadInputTokens\":0,\"cacheWriteInputTokenCount\":0,\"cacheWriteInputTokens\":0,\"inputTokens\":1150,\"outputTokens\":116,\"serverToolUsage\":{},\"totalTokens\":1266}}"
     headers:
       Connection:
       - keep-alive
       Content-Length:
-      - '246'
+      - '603'
       Content-Type:
       - application/json
       Date:
-      - Thu, 12 Feb 2026 22:01:12 GMT
-      x-amzn-ErrorType:
-      - ValidationException:http://internal.amazon.com/coral/com.amazon.bedrock/
+      - Fri, 24 Apr 2026 20:49:39 GMT
       x-amzn-RequestId:
       - X-AMZN-REQUESTID-XXX
     status:
-      code: 400
-      message: Bad Request
+      code: 200
+      message: OK
 version: 1

lib/crewai/tests/llms/anthropic/test_anthropic.py

@@ -975,7 +975,7 @@ def test_anthropic_agent_kickoff_structured_output_with_tools():
         role="Calculator",
         goal="Perform calculations using available tools",
         backstory="You are a calculator assistant that uses tools to compute results.",
-        llm=LLM(model="anthropic/claude-3-5-haiku-20241022"),
+        llm=LLM(model="anthropic/claude-sonnet-4-6"),
         tools=[add_numbers],
         verbose=True,
     )

lib/crewai/tests/llms/bedrock/test_bedrock.py

@@ -952,7 +952,7 @@ def test_bedrock_agent_kickoff_structured_output_with_tools():
         role="Calculator",
         goal="Perform calculations using available tools",
         backstory="You are a calculator assistant that uses tools to compute results.",
-        llm=LLM(model="bedrock/anthropic.claude-3-sonnet-20240229-v1:0"),
+        llm=LLM(model="bedrock/us.anthropic.claude-sonnet-4-6"),
         tools=[add_numbers],
         verbose=True,
     )