Same root cause as previous commit — tests that construct AIAgent()
without base_url rely on provider-resolver fallback state that doesn't
exist in hermetic CI / shard-split runs. Previously hidden because
other tests in the same xdist worker happened to prime module state.
Covered by the previous fix: calls that passed api_key but not base_url.
This covers calls that pass NEITHER (model=... only) — test_streaming.py
especially (24 call sites). Plus test_860_dedup, test_compression_
persistence, test_create_openai_client_*, test_provider_parity.
One call site (test_none_base_url_passed_as_none) remains explicitly
unmodified — it asserts None/empty base_url behavior, so adding base_url
would defeat the test's intent.
Validation:
- tests/run_agent/: 760 passed, 0 failed (local)
- Matrix shard 3 subset: 3098 passed, 0 failed, 1m49s (local)
Two new tests in tests/run_agent/ that pin the user-visible invariant
behind AlexKucera's Discord report (2026-04-16): no matter how a future
keepalive / transport fix for #10324 plumbs sockets in, sequential
chats on the same AIAgent instance must all succeed.
test_create_openai_client_reuse.py (no network, runs in CI):
- test_second_create_does_not_wrap_closed_transport_from_first
back-to-back _create_openai_client calls must not hand the same
http_client (after an SDK close) to the second construction
- test_replace_primary_openai_client_survives_repeated_rebuilds
three sequential rebuilds via the real _replace_primary_openai_client
entrypoint must each install a live client
test_sequential_chats_live.py (opt-in, HERMES_LIVE_TESTS=1):
- test_three_sequential_chats_across_client_rebuild
real OpenRouter round trips, with an explicit
_replace_primary_openai_client call between turns 2 and 3.
Error-sentinel detector treats 'API call failed after 3 retries'
replies as failures instead of letting them pass the naive
truthy check (which is how a first draft of this test missed
the bug it was meant to catch).
Validation:
clean main (post-revert, defensive copy present)
-> all 4 tests PASS
broken #10933 state (keepalive injection, no defensive copy)
-> all 4 tests FAIL with precise messages pointing at #10933
Companion to taeuk178's test_create_openai_client_kwargs_isolation.py,
which pins the syntactic 'don't mutate input dict' half of the same
contract. Together they catch both the specific mechanism of #10933
and any other reimplementation that breaks the sequential-call
invariant.
