The suite is 12.1k tests in a single CI job — nearly six minutes of pytest
that every push and every PR waits on in one block, on top of a setup step
that already installs npm, Playwright, FFmpeg and bubblewrap. Split it into
four sections the matrix runs in parallel: 352s becomes a 98s longest pole
locally.
Shards partition by test *file*, not by the area_* taxonomy markers. Those
markers do not partition the suite - a file can carry a hand-applied area_*
mark on top of the one conftest derives from its filename, so a marker-based
split would run those tests in more than one section. Assignment is a total
function of the file path instead, so every file lands in exactly one shard
and the four together run every test exactly once.
Sharding deselects rather than narrowing collection, so every test module is
still imported, in the same order, in every shard. The import-time stubbing
in conftest and the session-scoped static server behave identically whether
the suite runs whole or in sections - this suite has known collection-order
coupling and splitting by path would have walked into it.
Balance uses the existing `slow` marker as its weight signal rather than a
committed duration table that would go stale unnoticed. Files pack
heaviest-first into the lightest shard, which is deterministic for a given
file set, so every parallel job computes the same plan from the same commit.
Verified: the four shards together reproduce the full run exactly - 12141
tests selected across the four, and the same 59 failures, 78 skips and 2
xfails, by node ID and not merely by count.
The first checkpoint removed terminal-metrics discovery, but a normal
compact chat turn still ran two context systems: build_chat_context's
legacy untyped lookup (directly or inside maybe_compact) and the typed
resolver inside stream_preview.
Resolve the typed ContextResolution once, at the chat route, before
build_chat_context, using the session's provider credentials. The
predicate mirrors _clean_v3_preview; every input it needs is known at
that point and the native-workspace term cannot veto a requested clean
route. The same object then:
- sizes legacy history shaping in build_chat_context through a new
maybe_compact(context_length=...) override, so no legacy probe runs;
an unknown window still shapes with DEFAULT_CONTEXT but gains no
provenance;
- crosses stream_agent_loop (one new parameter, forwarded only at the
compact dispatch) into stream_preview, which reuses it and probes only
for callers that arrive without one or with one bound to another
route.
ContextResolution now records the endpoint and model it describes
(endpoint URL excluded from repr and metrics). The bare legacy
context_length is never converted into typed evidence.
Credential scoping: origins compare with default ports normalized, an
empty host is never trusted, and the probe client never follows
redirects. Tests cover the configured origin, the server-resolved
Tailscale form, scheme/port/lookalike/userinfo/path origins, redirects,
and secret-free errors, logs and metrics.
The conftest guard now replaces only the resolver's I/O edges (HTTP
client and DNS-capable URL building) instead of the whole probe, and
exposes a context_probe_ledger fixture, so route integration tests run
the real resolver offline and can count metadata requests.
The compact (clean v3) runtime had no effective context window: it learned a
limit only reactively from a provider 400/413 and its terminal metrics carried
no context_length. PR #41 addressed the reporting gap by probing provider
metadata between the last model byte and [DONE], unauthenticated, and folded
known-table and endpoint evidence into one "known" flag.
Resolve the window once, before the first model request, instead:
- src/agent_runtime/context_resolution.py adds a typed ContextResolution
(effective value, evidence class, source, all observations, conflicts,
provider_io, cached, secret-free probe errors). Evidence classes stay
distinct: runtime_confirmed (llama.cpp /slots, /props, or a limit the
provider stated this turn), provider_advertised (models catalog),
operator_declared (client_runtime_context.model_context_window),
known_table, unknown (0, never a default).
- Selection is deterministic: runtime beats provider beats table; an
operator declaration caps measured evidence and replaces weaker evidence.
Disagreements are recorded as conflicts; a declaration below a measured
value is a cap, above it a contradiction.
- The provider probe forwards the turn's credentials only to the provider's
own origin, runs URL resolution off the event loop, is bounded by one
deadline, never raises, and caches remote results per credential
fingerprint (shorter TTL for failures; local servers are re-probed).
- stream_preview resolves at preparation (or accepts a supplied resolution),
seeds the proactive trim budget from it when evidence is not unknown, and
terminal metrics report only the stored resolution plus any limit the
provider stated during the turn. Metrics perform no discovery.
src/agent_loop.py and the regular runtime's legacy model_context probe are
unchanged. A conftest guard keeps tests that drive the compact runtime with
placeholder endpoints from performing real DNS/HTTP lookups.
#25 fixed two sys.modules writes in test_auth_regressions.py that left empty
stub modules behind for the rest of the session, breaking 23 tests under one
collection order while the full suite stayed green. The class is wider than
that file, and an audit is the wrong answer to it: nothing stops the next one,
and the failure it causes lands on an unrelated test in a different file.
So this is a guard instead. An autouse fixture in the root conftest snapshots
which src.* / core.* names are bound to a bare ModuleType, and fails any test
that adds one. "Bare" is the same test the clear_fake_* helpers already use -
a plain types.ModuleType with no on-disk __file__. MagicMock stand-ins are out
of scope: they answer every attribute, so they fail at the point of use rather
than silently, and several files install them deliberately.
Three details that matter:
- It lives in the root conftest, so it is set up before any test-module
fixture and torn down after all of them. A stub a test's own teardown
removes is not reported.
- It drops the leaked entries as well as reporting them, so the failure stays
on the test that introduced it instead of cascading through the rest of the
run.
- It only reports stubs added during the test. Import state the session starts
with, including this conftest's own src.database stub, is left alone.
It found one beyond #25 on the first full run: _stub_heavy in
test_scheduler_restart_doublefire.py leaks the same five src.* modules as the
test #25 fixed, via sys.modules.setdefault. It already receives monkeypatch,
so the fix is to register through it. Fixed here because the guard has to land
green.
Full suite, macOS, default collection order:
this branch 10655 passed, 6 failed, 6 skipped 406s
lab 10655 passed, 6 failed, 6 skipped 371s
Same six either way, which is the point - none of this is visible in the
default order. Four are pre-existing macOS environment failures:
test_glob_confined_e2e and the two test_code_nav_tools document cases resolve
/tmp to /private/tmp, and
test_real_socket_falls_back_from_dead_first_to_live_second is connect-refused
timing on real sockets. The other two are the rich-colour and ffmpeg items
from the same ledger, fixed on their own branches.
Not verified: Linux, and any collection order other than the default. The
guard is order-independent by construction - it compares before and after
within a single test - but I have only run the default order.
The session-scoped autouse fixture bound 127.0.0.1:7011 and raised when the
port was taken. Because it is autouse, that raise errored every collected
test rather than the browser ones: a second worktree running its own suite
produced 10,612 errors, none of them about the code under test. 7011 is also
the application's own default port, so the suite could not run while a local
instance was up.
Bind port 0 instead and publish the resulting origin as
ODYSSEUS_TEST_STATIC_ORIGIN. The browser tests shell out to node, which
inherits the environment, so the snippets read process.env rather than
hardcoding a port. ODYSSEUS_TEST_STATIC_PORT still pins one when something
outside pytest has to reach the server; that is the only path that can now
fail to bind, and it fails with a message that says so.
Two concurrent full runs from one checkout now both pass. Only the docx export
snippet is an rf-string, so it is the only one whose JS braces needed doubling.
* docs: add implementation plan for fixing chat context drifting (#135)
* fix: make Session.history immutable + fix {}.history crash
- Session.history now exposes a COPY of the internal _history list
- add_message() replaces history with a fresh copy each time
- get_context_messages() derives from _history directly
- replace_messages() updates both _history and history
- truncate_messages() updates both _history and history
- _persist_message() line 207: fixed {}.history fallback crash
- Added 11 tests for session isolation and edge cases
Addresses #135 root cause #1: shared mutable references
* fix: task scheduler uses SessionManager methods instead of overwriting sessions
- Added ensure_task_session() to SessionManager (checks cache first)
- Task scheduler now uses ensure_task_session() instead of direct dict assignment
- Task scheduler now uses SessionManager.add_message() for message persistence
- Removed direct sess_obj.history.append() that was silently losing data
Addresses #135 root causes #2 and #3
* fix: add age guard to cleanup_empty_sessions — don't delete sessions <1h old
Prevents the cleanup task from deleting sessions that were just created
and haven't received any messages yet (message_count == 0).
Addresses #135 root cause #5
* test: comprehensive session isolation tests (10/10 passing)
* refactor: consolidate _session_manager into singleton pattern
- Added set_session_manager_instance / get_session_manager_instance to core/models
- kept backward-compat aliases (set_session_manager, get_session_manager)
- session_manager.py re-exports the singleton functions
- ai_interaction.set_session_manager now syncs with the core singleton
- context_compactor uses get_session_manager_instance() instead of getattr hack
- app.py initializes the singleton once
Addresses #135 root cause #4: fragile global wiring
* test: add concurrent session isolation integration tests
Verifies:
- Concurrent add_message to different sessions doesn't cross-contaminate
- Rapid parallel writes maintain isolation
- Read-write concurrent access is safe
All 3 async tests pass, proving the immutable history fix works under concurrency
* fix: pre-import core.models in conftest to prevent test pollution
test_agent_loop.py stubs sys.modules['core.models'] = MagicMock() at
module level during collection. Any test collected after it imports
Session as a MagicMock. Pre-importing core.models in conftest.py
before test_agent_loop.py's module-level code runs prevents this.
* fix: make .history authoritative mutable list, address PR review
Per review feedback: keep .history as the authoritative mutable list so
existing code doing .history.pop(), .history = [...], etc. still works.
Fix the cross-contamination bug by ensuring __post_init__() gives each
Session its OWN unique history list (never shared).
Changes:
- core/models.py: .history IS the authoritative list. _history aliases it.
Each Session gets its own list in __post_init__.
- core/session_manager.py: add_message() delegates to Session.add_message()
instead of appending directly — no double-append, single source of truth.
- tests/test_session_manager.py: updated test to reflect that .history
references see new messages (same list, not a snapshot).
- docs/plans/2026-06-01-fix-chat-context-drifting.md: removed (not for
shipping — useful design context but too much process/doc to ship).
All 272 tests pass (3 pre-existing failures unrelated).
* Fix session manager message persistence
* Fix session history alias regressions
* Fix session history aliasing and task delivery
Some test files (e.g. test_llm_core_sanitize_tool_calls) stub
sqlalchemy and core.database at module level with
`if mod not in sys.modules`. During pytest collection these stubs
fire before the real modules are imported, contaminating every
subsequent test that needs real ORM objects (IntegrityError, missing
columns, etc.).
Pre-import the real modules in conftest.py so the module-level
guards find them already loaded and skip the stubs. Fixes ~10+
cascading test failures that only appear in the full suite.
Fixes#2395
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Léo