mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
169 lines
4.7 KiB
Python
169 lines
4.7 KiB
Python
import sys
|
|
from types import ModuleType, SimpleNamespace
|
|
|
|
import pytest
|
|
from acp.schema import TextContentBlock
|
|
|
|
from acp_adapter.server import HermesACPAgent
|
|
from acp_adapter.session import SessionManager
|
|
|
|
|
|
class FakeAgent:
|
|
def __init__(self):
|
|
self.model = "fake-model"
|
|
self.provider = "fake-provider"
|
|
self.enabled_toolsets = ["hermes-acp"]
|
|
self.disabled_toolsets = []
|
|
self.tools = []
|
|
self.valid_tool_names = set()
|
|
self._supports_active_turn_redirect = True
|
|
self.steers = []
|
|
self.redirects = []
|
|
self.runs = []
|
|
|
|
def steer(self, text):
|
|
self.steers.append(text)
|
|
return True
|
|
|
|
def redirect(self, text):
|
|
self.redirects.append(text)
|
|
return True
|
|
|
|
def run_conversation(self, *, user_message, conversation_history, task_id, **kwargs):
|
|
self.runs.append(user_message)
|
|
messages = list(conversation_history or [])
|
|
messages.append({"role": "user", "content": user_message})
|
|
final = f"ran: {user_message}"
|
|
messages.append({"role": "assistant", "content": final})
|
|
return {"final_response": final, "messages": messages}
|
|
|
|
|
|
class CaptureConn:
|
|
def __init__(self):
|
|
self.updates = []
|
|
|
|
async def session_update(self, *args, **kwargs):
|
|
if kwargs:
|
|
self.updates.append((kwargs.get("session_id"), kwargs.get("update")))
|
|
else:
|
|
self.updates.append((args[0], args[1]))
|
|
|
|
async def request_permission(self, *args, **kwargs):
|
|
return SimpleNamespace(outcome="allow")
|
|
|
|
|
|
class NoopDb:
|
|
def get_session(self, *_args, **_kwargs):
|
|
return None
|
|
|
|
def create_session(self, *_args, **_kwargs):
|
|
return None
|
|
|
|
def update_session(self, *_args, **_kwargs):
|
|
return None
|
|
|
|
|
|
def make_agent_and_state():
|
|
fake = FakeAgent()
|
|
manager = SessionManager(agent_factory=lambda **kwargs: fake, db=NoopDb())
|
|
acp_agent = HermesACPAgent(session_manager=manager)
|
|
state = manager.create_session(cwd=".")
|
|
conn = CaptureConn()
|
|
acp_agent.on_connect(conn)
|
|
return acp_agent, state, fake, conn
|
|
|
|
|
|
def test_acp_real_agent_gets_session_db_for_recall(monkeypatch):
|
|
"""ACP sessions persist to SessionDB; recall must receive the same DB handle."""
|
|
captured = {}
|
|
sentinel_db = NoopDb()
|
|
|
|
class CapturingAgent(FakeAgent):
|
|
def __init__(self, **kwargs):
|
|
super().__init__()
|
|
captured.update(kwargs)
|
|
|
|
def mod(name, **attrs):
|
|
module = ModuleType(name)
|
|
for key, value in attrs.items():
|
|
setattr(module, key, value)
|
|
return module
|
|
|
|
monkeypatch.setitem(sys.modules, "run_agent", mod("run_agent", AIAgent=CapturingAgent))
|
|
monkeypatch.setitem(
|
|
sys.modules,
|
|
"hermes_cli.config",
|
|
mod("hermes_cli.config", load_config=lambda: {"model": {"default": "m", "provider": "p"}}),
|
|
)
|
|
monkeypatch.setitem(
|
|
sys.modules,
|
|
"hermes_cli.runtime_provider",
|
|
mod(
|
|
"hermes_cli.runtime_provider",
|
|
resolve_runtime_provider=lambda **_kwargs: {
|
|
"provider": "p",
|
|
"api_mode": "chat_completions",
|
|
"base_url": "u",
|
|
"api_key": "k",
|
|
"command": None,
|
|
"args": [],
|
|
},
|
|
),
|
|
)
|
|
|
|
manager = SessionManager(db=sentinel_db)
|
|
agent = manager._make_agent(session_id="acp-session", cwd=".")
|
|
|
|
assert isinstance(agent, CapturingAgent)
|
|
assert captured["session_db"] is sentinel_db
|
|
assert captured["platform"] == "acp"
|
|
assert captured["session_id"] == "acp-session"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_acp_steer_slash_command_injects_into_running_agent():
|
|
acp_agent, state, fake, _conn = make_agent_and_state()
|
|
state.is_running = True
|
|
|
|
response = await acp_agent.prompt(
|
|
session_id=state.session_id,
|
|
prompt=[TextContentBlock(type="text", text="/steer prefer the simpler fix")],
|
|
)
|
|
|
|
assert response.stop_reason == "end_turn"
|
|
assert fake.steers == ["prefer the simpler fix"]
|
|
assert fake.runs == []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_acp_cancel_publishes_hard_stop_while_holding_runtime_lock():
|
|
acp_agent, state, fake, _conn = make_agent_and_state()
|
|
state.is_running = True
|
|
state.current_prompt_text = "original request"
|
|
observed = {}
|
|
|
|
def interrupt():
|
|
acquired = state.runtime_lock.acquire(blocking=False)
|
|
observed["lock_held"] = not acquired
|
|
if acquired:
|
|
state.runtime_lock.release()
|
|
|
|
fake.interrupt = interrupt
|
|
|
|
await acp_agent.cancel(state.session_id)
|
|
|
|
assert observed["lock_held"] is True
|
|
assert state.cancel_event.is_set()
|
|
assert state.interrupted_prompt_text == "original request"
|
|
|
|
|
|
|
|
|
|
|
|
|