fix(heartbeat): inject current datetime into Phase 1 prompt
Phase 1 _decide() now includes "Current date/time: YYYY-MM-DD HH:MM UTC" in the user prompt and instructs the LLM to use it for time-aware scheduling. Without this, the LLM defaults to 'run' for any task description regardless of whether it is actually due, defeating Phase 1's pre-screening purpose. Closes #1929
This commit is contained in:
@@ -3,6 +3,7 @@
|
|||||||
from __future__ import annotations
|
from __future__ import annotations
|
||||||
|
|
||||||
import asyncio
|
import asyncio
|
||||||
|
from datetime import datetime, timezone
|
||||||
from pathlib import Path
|
from pathlib import Path
|
||||||
from typing import TYPE_CHECKING, Any, Callable, Coroutine
|
from typing import TYPE_CHECKING, Any, Callable, Coroutine
|
||||||
|
|
||||||
@@ -87,10 +88,19 @@ class HeartbeatService:
|
|||||||
|
|
||||||
Returns (action, tasks) where action is 'skip' or 'run'.
|
Returns (action, tasks) where action is 'skip' or 'run'.
|
||||||
"""
|
"""
|
||||||
|
now_str = datetime.now(timezone.utc).strftime("%Y-%m-%d %H:%M UTC")
|
||||||
|
|
||||||
response = await self.provider.chat_with_retry(
|
response = await self.provider.chat_with_retry(
|
||||||
messages=[
|
messages=[
|
||||||
{"role": "system", "content": "You are a heartbeat agent. Call the heartbeat tool to report your decision."},
|
{"role": "system", "content": (
|
||||||
|
"You are a heartbeat agent. Call the heartbeat tool to report your decision. "
|
||||||
|
"The current date/time is provided so you can evaluate time-based conditions. "
|
||||||
|
"Choose 'run' if there are active tasks to execute. "
|
||||||
|
"Choose 'skip' if the file has no actionable tasks, if blocking conditions "
|
||||||
|
"are not yet met, or if tasks are scheduled for a future time that has not arrived yet."
|
||||||
|
)},
|
||||||
{"role": "user", "content": (
|
{"role": "user", "content": (
|
||||||
|
f"Current date/time: {now_str}\n\n"
|
||||||
"Review the following HEARTBEAT.md and decide whether there are active tasks.\n\n"
|
"Review the following HEARTBEAT.md and decide whether there are active tasks.\n\n"
|
||||||
f"{content}"
|
f"{content}"
|
||||||
)},
|
)},
|
||||||
|
|||||||
@@ -250,3 +250,47 @@ async def test_decide_retries_transient_error_then_succeeds(tmp_path, monkeypatc
|
|||||||
assert tasks == "check open tasks"
|
assert tasks == "check open tasks"
|
||||||
assert provider.calls == 2
|
assert provider.calls == 2
|
||||||
assert delays == [1]
|
assert delays == [1]
|
||||||
|
|
||||||
|
|
||||||
|
@pytest.mark.asyncio
|
||||||
|
async def test_decide_prompt_includes_current_datetime(tmp_path) -> None:
|
||||||
|
"""Phase 1 prompt must contain the current date/time so the LLM can judge task urgency."""
|
||||||
|
|
||||||
|
captured_messages: list[dict] = []
|
||||||
|
|
||||||
|
class CapturingProvider(LLMProvider):
|
||||||
|
async def chat(self, *, messages=None, **kwargs) -> LLMResponse:
|
||||||
|
if messages:
|
||||||
|
captured_messages.extend(messages)
|
||||||
|
return LLMResponse(
|
||||||
|
content="",
|
||||||
|
tool_calls=[
|
||||||
|
ToolCallRequest(
|
||||||
|
id="hb_1", name="heartbeat",
|
||||||
|
arguments={"action": "skip"},
|
||||||
|
)
|
||||||
|
],
|
||||||
|
)
|
||||||
|
|
||||||
|
def get_default_model(self) -> str:
|
||||||
|
return "test-model"
|
||||||
|
|
||||||
|
service = HeartbeatService(
|
||||||
|
workspace=tmp_path,
|
||||||
|
provider=CapturingProvider(),
|
||||||
|
model="test-model",
|
||||||
|
)
|
||||||
|
|
||||||
|
await service._decide("- [ ] check servers at 10:00 UTC")
|
||||||
|
|
||||||
|
# System prompt should mention date/time awareness
|
||||||
|
system_msg = captured_messages[0]
|
||||||
|
assert system_msg["role"] == "system"
|
||||||
|
assert "date/time" in system_msg["content"].lower()
|
||||||
|
|
||||||
|
# User prompt should contain a UTC timestamp
|
||||||
|
user_msg = captured_messages[1]
|
||||||
|
assert user_msg["role"] == "user"
|
||||||
|
assert "Current date/time:" in user_msg["content"]
|
||||||
|
assert "UTC" in user_msg["content"]
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user