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.
133 lines
5.1 KiB
Python
133 lines
5.1 KiB
Python
"""Regression tests for topic/channel skill auto-injection after /new or /reset.
|
|
|
|
Covers the fix for issue #6508.
|
|
|
|
Before the fix:
|
|
1. User sends ``/new`` — ``reset_session`` creates a fresh SessionEntry
|
|
with ``created_at == updated_at``.
|
|
2. User sends the next message.
|
|
3. ``get_or_create_session`` finds the entry and bumps
|
|
``entry.updated_at = now`` (microseconds after ``created_at``).
|
|
4. ``_handle_message_with_agent`` checks
|
|
``_is_new_session = (created_at == updated_at) or was_auto_reset``.
|
|
Both are False → ``_is_new_session = False`` → topic/channel skills
|
|
are silently skipped for the first message of a manually reset session.
|
|
|
|
After the fix:
|
|
``reset_session`` stamps the new entry with ``is_fresh_reset=True``.
|
|
``_handle_message_with_agent`` ORs this into ``_is_new_session`` and
|
|
consumes the flag immediately after the check, so subsequent messages
|
|
are treated as continuing the session and the flag does not leak.
|
|
|
|
We use ``was_auto_reset`` for surprise resets (idle/daily/suspended) and
|
|
``is_fresh_reset`` for user-initiated resets because the former also drives
|
|
a "Session automatically reset due to inactivity" user-facing notice and
|
|
a context-note prepend into the agent's prompt — both wrong for an explicit
|
|
/new or /reset.
|
|
"""
|
|
|
|
from gateway.config import GatewayConfig, Platform
|
|
from gateway.session import SessionEntry, SessionSource, SessionStore
|
|
|
|
|
|
def _make_store(tmp_path):
|
|
return SessionStore(sessions_dir=tmp_path, config=GatewayConfig())
|
|
|
|
|
|
def _make_source(chat_id="123", user_id="u1"):
|
|
return SessionSource(
|
|
platform=Platform.TELEGRAM,
|
|
chat_id=chat_id,
|
|
user_id=user_id,
|
|
)
|
|
|
|
|
|
def _is_new_session(entry) -> bool:
|
|
"""Mirror of the predicate in ``_handle_message_with_agent``.
|
|
|
|
Kept in-sync with the production check so this test fails loudly if the
|
|
upstream logic regresses.
|
|
"""
|
|
return (
|
|
entry.created_at == entry.updated_at
|
|
or getattr(entry, "was_auto_reset", False)
|
|
or getattr(entry, "is_fresh_reset", False)
|
|
)
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# reset_session stamps is_fresh_reset=True
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestResetSessionStampsFreshReset:
|
|
def test_reset_session_sets_is_fresh_reset_true(self, tmp_path):
|
|
store = _make_store(tmp_path)
|
|
source = _make_source()
|
|
store.get_or_create_session(source)
|
|
session_key = store._generate_session_key(source)
|
|
|
|
new_entry = store.reset_session(session_key)
|
|
|
|
assert new_entry is not None
|
|
assert new_entry.is_fresh_reset is True
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Core regression: _is_new_session stays True after updated_at bump
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestIsNewSessionSurvivesUpdatedAtBump:
|
|
def test_is_new_session_true_after_reset_then_next_message(self, tmp_path):
|
|
"""The actual bug: _is_new_session was False on message after /reset."""
|
|
store = _make_store(tmp_path)
|
|
source = _make_source()
|
|
store.get_or_create_session(source)
|
|
session_key = store._generate_session_key(source)
|
|
|
|
# User sends /reset
|
|
store.reset_session(session_key)
|
|
|
|
# Next inbound message — get_or_create_session bumps updated_at
|
|
entry = store.get_or_create_session(source)
|
|
|
|
# Before the fix: created_at != updated_at, was_auto_reset=False → False
|
|
# After the fix: is_fresh_reset=True carries the signal through the bump
|
|
assert _is_new_session(entry) is True
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Vanilla-session behavior is unchanged
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestVanillaBehaviorUnaffected:
|
|
def test_ongoing_session_not_flagged_as_new(self, tmp_path):
|
|
store = _make_store(tmp_path)
|
|
source = _make_source()
|
|
store.get_or_create_session(source)
|
|
|
|
# Second message on the same session — updated_at bumps,
|
|
# is_fresh_reset was never set
|
|
entry = store.get_or_create_session(source)
|
|
assert entry.is_fresh_reset is False
|
|
assert _is_new_session(entry) is False
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Persistence through sessions.json round-trip
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestPersistence:
|
|
def test_is_fresh_reset_survives_to_dict_from_dict(self, tmp_path):
|
|
"""Protect against the gateway restarting between /reset and the
|
|
next message — the flag must be persisted in sessions.json.
|
|
"""
|
|
store = _make_store(tmp_path)
|
|
source = _make_source()
|
|
store.get_or_create_session(source)
|
|
session_key = store._generate_session_key(source)
|
|
new_entry = store.reset_session(session_key)
|
|
|
|
assert new_entry.is_fresh_reset is True
|
|
restored = SessionEntry.from_dict(new_entry.to_dict())
|
|
assert restored.is_fresh_reset is True
|
|
|