Removed tool_names validation for gen_answer being present

paperqa/agents/env.py

@@ -47,7 +47,6 @@ def settings_to_tools(
     summary_llm_model = summary_llm_model or settings.get_summary_llm()
     embedding_model = embedding_model or settings.get_embedding_model()
     tools: list[Tool] = []
-    has_answer_tool = False
     for tool_type in (
         (PaperSearch, GatherEvidence, GenerateAnswer)
         if settings.agent.tool_names is None
@@ -86,14 +85,9 @@ def settings_to_tools(
         else:
             raise NotImplementedError(f"Didn't handle tool type {tool_type}.")
         if tool.info.name == GenerateAnswer.gen_answer.__name__:
-            tools.append(tool)
-            has_answer_tool = True
+            tools.append(tool)  # Place at the end
         else:
             tools.insert(0, tool)
-    if not has_answer_tool:
-        raise ValueError(
-            f"{GenerateAnswer.gen_answer.__name__} must be one of the tools."
-        )
     return tools
 
 

paperqa/settings.py

@@ -523,21 +523,6 @@ class AgentSettings(BaseModel):
         exclude=True,
     )
 
-    @field_validator("tool_names")
-    @classmethod
-    def validate_tool_names(cls, v: set[str] | None) -> set[str] | None:
-        if v is None:
-            return None
-        # imported here to avoid circular imports
-        from paperqa.agents.tools import GenerateAnswer
-
-        answer_tool_name = GenerateAnswer.TOOL_FN_NAME
-        if answer_tool_name not in v:
-            raise ValueError(
-                f"If using an override, must contain at least the {answer_tool_name}."
-            )
-        return v
-
     @model_validator(mode="after")
     def _deprecated_field(self) -> Self:
         for deprecated_field_name, new_name in (("index_concurrency", "concurrency"),):

tests/test_agents.py

@@ -10,7 +10,7 @@
 import time
 from copy import deepcopy
 from pathlib import Path
-from typing import Any, cast
+from typing import cast
 from unittest.mock import AsyncMock, patch
 from uuid import uuid4
 
@@ -21,7 +21,6 @@
 from ldp.graph.memory import Memory, UIndexMemoryModel
 from ldp.graph.ops import OpResult
 from ldp.llms import EmbeddingModel, MultipleCompletionLLMModel
-from pydantic import ValidationError
 from pytest_subtests import SubTests
 from tantivy import Index
 
@@ -402,8 +401,8 @@ async def test_propagate_options(agent_test_settings: Settings) -> None:
 async def test_gather_evidence_rejects_empty_docs(
     agent_test_settings: Settings,
 ) -> None:
-    # Patch GenerateAnswerTool._arun so that if this tool is chosen first, we
-    # don't give a 'cannot answer' response. A 'cannot answer' response can
+    # Patch GenerateAnswerTool.gen_answer so that if this tool is chosen first,
+    # we don't give a 'cannot answer' response. A 'cannot answer' response can
     # lead to an unsure status, which will break this test's assertions. Since
     # this test is about a GatherEvidenceTool edge case, defeating
     # GenerateAnswerTool is fine
@@ -726,25 +725,6 @@ def test_answers_are_striped() -> None:
     response.model_dump_json()
 
 
-@pytest.mark.parametrize(
-    ("kwargs", "result"),
-    [
-        ({}, None),
-        ({"tool_names": {GenerateAnswer.TOOL_FN_NAME}}, None),
-        ({"tool_names": set()}, ValidationError),
-        ({"tool_names": {PaperSearch.TOOL_FN_NAME}}, ValidationError),
-    ],
-)
-def test_agent_prompt_collection_validations(
-    kwargs: dict[str, Any], result: type[Exception] | None
-) -> None:
-    if result is None:
-        AgentSettings(**kwargs)
-    else:
-        with pytest.raises(result):
-            AgentSettings(**kwargs)
-
-
 class TestGradablePaperQAEnvironment:
     @pytest.mark.flaky(reruns=2, only_rerun=["AssertionError"])
     @pytest.mark.asyncio