Update unknown tool retry logic

pydantic_ai_slim/pydantic_ai/_tool_manager.py

@@ -35,6 +35,8 @@ class ToolManager(Generic[AgentDepsT]):
     """The cached tools for this run step."""
     failed_tools: set[str] = field(default_factory=set)
     """Names of tools that failed in this run step."""
+    default_max_retries: int = 1
+    """Default number of times to retry a tool"""
 
     @classmethod
     @contextmanager
@@ -62,6 +64,7 @@ async def for_run_step(self, ctx: RunContext[AgentDepsT]) -> ToolManager[AgentDe
             toolset=self.toolset,
             ctx=ctx,
             tools=await self.toolset.get_tools(ctx),
+            default_max_retries=self.default_max_retries,
         )
 
     @property
@@ -174,7 +177,7 @@ async def _call_tool(
 
             return await self.toolset.call_tool(name, args_dict, ctx, tool)
         except (ValidationError, ModelRetry) as e:
-            max_retries = tool.max_retries if tool is not None else 1
+            max_retries = tool.max_retries if tool is not None else self.default_max_retries
             current_retry = self.ctx.retries.get(name, 0)
 
             if current_retry == max_retries:

pydantic_ai_slim/pydantic_ai/agent/__init__.py

@@ -589,7 +589,7 @@ async def main():
                 output_toolset.max_retries = self._max_result_retries
                 output_toolset.output_validators = output_validators
         toolset = self._get_toolset(output_toolset=output_toolset, additional_toolsets=toolsets)
-        tool_manager = ToolManager[AgentDepsT](toolset)
+        tool_manager = ToolManager[AgentDepsT](toolset, default_max_retries=self._max_tool_retries)
 
         # Build the graph
         graph = _agent_graph.build_agent_graph(self.name, self._deps_type, output_type_)

tests/test_agent.py

@@ -2765,6 +2765,73 @@ def empty(m: list[ModelMessage], _info: AgentInfo) -> ModelResponse:
     )
 
 
+def test_unknown_tool_multiple_retries():
+    num_retries = 2
+
+    def empty(_: list[ModelMessage], _info: AgentInfo) -> ModelResponse:
+        return ModelResponse(parts=[ToolCallPart('foobar', '{}')])
+
+    def foo() -> str:
+        return 'Hello from foo'
+
+    agent = Agent(FunctionModel(empty), tools=[foo], retries=num_retries)
+
+    with capture_run_messages() as messages:
+        with pytest.raises(UnexpectedModelBehavior, match=r'Exceeded maximum retries \(2\) for output validation'):
+            agent.run_sync('Hello')
+    assert messages == snapshot(
+        [
+            ModelRequest(
+                parts=[UserPromptPart(content='Hello', timestamp=IsNow(tz=timezone.utc))],
+                run_id=IsStr(),
+            ),
+            ModelResponse(
+                parts=[ToolCallPart(tool_name='foobar', args='{}', tool_call_id=IsStr())],
+                usage=RequestUsage(input_tokens=51, output_tokens=2),
+                model_name='function:empty:',
+                timestamp=IsNow(tz=timezone.utc),
+                run_id=IsStr(),
+            ),
+            ModelRequest(
+                parts=[
+                    RetryPromptPart(
+                        tool_name='foobar',
+                        content="Unknown tool name: 'foobar'. Available tools: 'foo'",
+                        tool_call_id=IsStr(),
+                        timestamp=IsNow(tz=timezone.utc),
+                    )
+                ],
+                run_id=IsStr(),
+            ),
+            ModelResponse(
+                parts=[ToolCallPart(tool_name='foobar', args='{}', tool_call_id=IsStr())],
+                usage=RequestUsage(input_tokens=65, output_tokens=4),
+                model_name='function:empty:',
+                timestamp=IsNow(tz=timezone.utc),
+                run_id=IsStr(),
+            ),
+            ModelRequest(
+                parts=[
+                    RetryPromptPart(
+                        tool_name='foobar',
+                        content="Unknown tool name: 'foobar'. Available tools: 'foo'",
+                        tool_call_id=IsStr(),
+                        timestamp=IsNow(tz=timezone.utc),
+                    )
+                ],
+                run_id=IsStr(),
+            ),
+            ModelResponse(
+                parts=[ToolCallPart(tool_name='foobar', args='{}', tool_call_id=IsStr())],
+                usage=RequestUsage(input_tokens=79, output_tokens=6),
+                model_name='function:empty:',
+                timestamp=IsNow(tz=timezone.utc),
+                run_id=IsStr(),
+            ),
+        ]
+    )
+
+
 def test_tool_exceeds_token_limit_error():
     def return_incomplete_tool(_: list[ModelMessage], info: AgentInfo) -> ModelResponse:
         resp = ModelResponse(parts=[ToolCallPart('dummy_tool', args='{"foo": "bar",')])