mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
39975613b1
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.
54 lines
2.2 KiB
Python
54 lines
2.2 KiB
Python
"""Regression tests for PR #48127: cached agent max_iterations refresh.
|
|
|
|
When a long-lived gateway reuses an agent from its cache, the agent must run
|
|
the *current* configured iteration budget — not the budget it was constructed
|
|
with on the first turn of that session. Two pieces make that true:
|
|
|
|
1. ``GatewayRunner._init_cached_agent_for_turn`` must NOT reset
|
|
``max_iterations`` itself (the gateway refreshes it explicitly right after,
|
|
from current config). If this helper ever started clobbering it, the
|
|
gateway's refresh would be silently undone.
|
|
2. The per-turn budget object is rebuilt from ``agent.max_iterations`` at the
|
|
start of every turn (``agent/turn_context.py`` -> ``IterationBudget``), so
|
|
refreshing ``max_iterations`` on the cached agent is sufficient to change
|
|
the operative cap the agent loop checks.
|
|
|
|
These tests exercise the real code paths rather than asserting a plain
|
|
assignment, so they fail if either contract regresses.
|
|
"""
|
|
|
|
import time
|
|
from types import SimpleNamespace
|
|
|
|
from agent.iteration_budget import IterationBudget
|
|
|
|
|
|
def _make_cached_agent(max_iterations: int) -> SimpleNamespace:
|
|
"""A minimal stand-in cached agent with the attributes the helpers touch."""
|
|
# The turn loop checks both api_call_count >= max_iterations AND
|
|
# iteration_budget.remaining <= 0 (turn_finalizer.py), so the budget must
|
|
# also reflect the new cap. Seed it with the stale value to prove the
|
|
# refresh propagates.
|
|
return SimpleNamespace(
|
|
_last_activity_ts=time.time() - 1000,
|
|
_last_activity_desc="previous turn",
|
|
_api_call_count=42,
|
|
_last_flushed_db_idx=5,
|
|
max_iterations=max_iterations,
|
|
iteration_budget=IterationBudget(max_iterations),
|
|
)
|
|
|
|
|
|
def test_init_cached_agent_preserves_max_iterations_on_interrupt_depth():
|
|
"""Interrupt-recursive turns must also leave max_iterations alone."""
|
|
from gateway.run import GatewayRunner
|
|
|
|
agent = _make_cached_agent(200)
|
|
GatewayRunner._init_cached_agent_for_turn(agent, interrupt_depth=1)
|
|
|
|
# Activity timestamps preserved for the inactivity watchdog (#15654)...
|
|
assert agent._last_activity_desc == "previous turn"
|
|
# ...and max_iterations untouched.
|
|
assert agent.max_iterations == 200
|
|
|
|
|