ece8a5ebed
model definitions now support a new `allow_code_generation` flag, only to be used with higher reasoning models such as GPT-5-Pro and-Gemini 2.5-Pro When `true`, the `chat` tool can now request the external model to generate a full implementation / update / instructions etc and then share the implementation with the calling agent. This effectively allows us to utilize more powerful models such as GPT-5-Pro to generate code for us or entire implementations (which are either API-only or part of the $200 Pro plan from within the ChatGPT app)
37 lines
1.4 KiB
Python
37 lines
1.4 KiB
Python
"""Integration test for conversation continuation persistence."""
|
|
|
|
from tools.chat import ChatRequest, ChatTool
|
|
from utils.conversation_memory import get_thread
|
|
from utils.storage_backend import get_storage_backend
|
|
|
|
|
|
def test_first_response_persisted_in_conversation_history(tmp_path):
|
|
"""Ensure the assistant's initial reply is stored for newly created threads."""
|
|
|
|
# Clear in-memory storage to avoid cross-test contamination
|
|
storage = get_storage_backend()
|
|
storage._store.clear() # type: ignore[attr-defined]
|
|
|
|
tool = ChatTool()
|
|
request = ChatRequest(prompt="First question?", model="local-llama", working_directory=str(tmp_path))
|
|
response_text = "Here is the initial answer."
|
|
|
|
# Mimic the first tool invocation (no continuation_id supplied)
|
|
continuation_data = tool._create_continuation_offer(request, model_info={"model_name": "local-llama"})
|
|
tool._create_continuation_offer_response(
|
|
response_text,
|
|
continuation_data,
|
|
request,
|
|
{"model_name": "local-llama", "provider": "custom"},
|
|
)
|
|
|
|
thread_id = continuation_data["continuation_id"]
|
|
thread = get_thread(thread_id)
|
|
|
|
assert thread is not None
|
|
assert [turn.role for turn in thread.turns] == ["user", "assistant"]
|
|
assert thread.turns[-1].content == response_text
|
|
|
|
# Cleanup storage for subsequent tests
|
|
storage._store.clear() # type: ignore[attr-defined]
|