After one failed/stalled summary attempt arms the 60/300/900s compression-
failure cooldown, a provider context_length_exceeded rejection entered the
reactive overflow branch in conversation_loop, which called _compress_context
without force. Since #97488 the cooldown gate returns the soft "temporarily
paused, retry in a moment" deferral instead of exhaustion, so every turn
deferred until the cooldown lapsed, and the next failure extended the ladder:
long-running sessions wedged with no automatic recovery (#100661, four sessions
lost).
Thread a narrow `bypass_cooldown` kwarg from the three provider-proven overflow
call sites (generic overflow, 413, output-cap recovery) through
AIAgent._compress_context -> compress_context -> ContextCompressor.compress ->
_generate_summary. It skips ONLY the summary-failure cooldown check at each gate.
Unlike force=True it does not clear the cooldown, does not skip the feasibility /
anti-thrash breakers, and a failed attempt records its cooldown normally. The
attempt is bounded by the existing compression_attempts/max_compression_attempts
budget, so there is no retry loop. The preflight threshold gate is unchanged:
ordinary over-threshold pressure still honors the cooldown (#11529).
Engines whose _automatic_compression_blocked()/compress() predate the kwarg
(plugins, test doubles) are called with the legacy signature.
Tests: cooldown armed + bypass_cooldown -> summarizer invoked and transcript
compacted; ordinary pass still deferred. Docs note the cooldown/overflow
contract in the developer guide.
Fixes#100661Closes#97766 (overflow-force idea; the bundled continuation changes were not taken)
Co-authored-by: sgtworkman <178342791+sgtworkman@users.noreply.github.com>
A bare SQLITE_CORRUPT/NOTADB on a live write (not FTS-scoped, not a
replaced file) now sets a sticky per-instance flag: later writes fail
fast with StateDbCorruptError, the handle never reopens after close(),
and close() skips its explicit PASSIVE WAL checkpoint. Gateway and agent
flush paths divert pending transcripts to JSONL/spool like the replaced
case instead of retrying forever.
Field evidence: a handle that kept writing for ~50 minutes after the
first structural error checkpointed 15 pages under the wrong page
numbers on shutdown (page 1 <- messages_fts_trigram_data leaf), turning
"malformed" into "file is not a database".
Refs #90837, #90950, #97940, #89332, #45383
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CNX8rNYHqA5pT4tAGSzXtb
Move the structural clone from the four call sites (auto review, codex
runtime, CLI /refine, gateway /refine) into AIAgent._spawn_background_review,
which every review path — immediate, idle-queue deferred, requeued — passes
through. Callers can no longer forget it, and the private helper is no longer
imported across hermes_cli/ and gateway/ package boundaries.
Tests now bind the real chokepoint (capturing at _spawn_background_review_now)
so they still fail if the clone is removed.
Widen the salvaged #49189 hard-stop default so it covers the loop shape in
the #100849 debug bundle and #89069: a model replaying the same SUCCESSFUL
call (terminal, skill_view, memory) with a byte-identical result. The
per-turn idempotent_no_progress block only tracks IDEMPOTENT_TOOL_NAMES, so
those loops ran until the iteration budget (600 calls, ~40 min) with only a
notice appended.
- agent/tool_guardrails.py: observe_call's tool-agnostic consecutive-identical
streak raises a halt (identical_call_streak_halt) at
hard_stop_after.idempotent_no_progress when hard stops are active. Pollers
stay exempt; a changed result resets the streak; warning-only sessions are
unchanged.
- run_agent.py: surface that halt from _append_guardrail_observation like
every other guardrail halt (appends guidance, ends the turn).
- hermes_cli/config_defaults.py: declare non_interactive_hard_stop_enabled.
- docs: configuration.md describes the streak hard-stop.
- tests: streak halts terminal under hard_stop; never under soft mode,
for pollers, or when results change.
Live A/B (real AIAgent platform=telegram, mocked client replaying one call):
identical failing read_file main: 602 API calls, budget exhausted
branch: 8 calls, repeated_exact_failure_block
identical successful terminal main: 602 API calls, budget exhausted
branch: 5 calls, identical_call_streak_halt
Ollama cloud models (model name contains ':cloud') run generation on
Ollama's hosted server; the local 11434 endpoint is only a transparent
proxy that forwards finish_reason faithfully. _is_ollama_glm_backend()
was matching these models because the proxy listens on the same port
as local Ollama, causing _should_treat_stop_as_truncated() to rewrite
a correct finish_reason='stop' into 'length'.
The 4-attempt continuation loop then injects a synthetic user nudge
that reasoning-capable GLM models spend their output budget deliberating
over, producing unpunctuated tails that re-trigger _has_natural_response_ending()
rejection -- a self-reinforcing loop that always exhausts retries.
Fix: add an early return in _is_ollama_glm_backend when the model name
contains ':cloud'. Local GLM inference (no ':cloud') is still caught
by the existing port/URL/provider checks.
Fixes#98406
Two compounding bugs that cause WebUI to discard or misrender agent
responses when using GLM models on Ollama Cloud:
1. _is_ollama_glm_backend() matched "ollama" in base URL, which
included Ollama Cloud (ollama.com). The hosted service correctly
reports finish_reason and is not affected by the local Ollama
stop-reason bug. Exclude "ollama.com" before the substring check.
2. _handle_session_chat_stream() hardcoded "partial": False in the
assistant.completed SSE event instead of reading result.get("partial").
The WebUI could not detect truncation and rendered partial responses
incorrectly (showing only the continuation instead of the full text).
Read the partial flag from the agent result, matching the pattern
used by other SSE paths in the same file.
Fixes#72316
On reasoning models a long tool loop replays the current turn's thinking +
scaffolding on every request, so the LAST request's prompt_tokens can exceed
the durable transcript by hundreds of K — all of which evaporates at the turn
boundary. The status bar and /context breakdown rendered that raw figure, so
users watched 'context' jump (e.g.) 850K -> 600K across a turn boundary and
read it as a broken compaction.
- conversation_loop: capture a turn-base usage anchor from the turn's FIRST
provider response (api_call_count == 1), where replay is minimal.
- anchored_context_tokens: new charge_stale_thinking kwarg forwarded to the
delta estimate (stale reasoning excluded on all but the newest assistant
message).
- cli status snapshot + context_breakdown: prefer the turn-base anchored
figure; fall back to last-response anchor / raw last_prompt_tokens.
- All _usage_anchor invalidation sites also clear _turn_base_usage_anchor.
Display-only: compression trigger math keeps using real last-request usage
(the inflated request is what actually risks the window mid-loop).
Run models locally as a first-class provider. The CLI grows a managed
llama.cpp runtime (engine install, model download, server supervision);
the desktop app grows the full setup and management story on top of it.
GUI surfaces ship behind the desktop --local launch flag (hermes desktop
--local, or the flag on the packaged app); backend routes and the CLI
are always live.
Runtime (hermes_cli/local_runtime/):
- curated GGUF catalog with per-machine variant selection: hardware
probe (VRAM/RAM/UMA), fit planning with spill accounting, quant choice
by context window
- derived recommendation: quality-ranked picks gated by a predicted
decode-speed floor, bandwidth-aware on unified memory; the decision
table is pinned as a test (pick AND reason per memory class), and the
Recommended badge explains its pick in a tooltip fed by the resolver's
actual branch
- engine install + model download with resumable split parts, cumulative
plan-level progress, and staged-model integrity (a split GGUF counts
only when every part is present)
- server supervision: spawn/adopt/stop, router mode with per-model load
progress relayed over SSE, abandoned-request cleanup
Desktop:
- Settings -> Providers -> Local models: one-click quickstart (install
engine, download the recommended model, boot) plus per-model download/
activate/eject, fit-ranked catalog with context pills
- model pickers (composer dropdown + Cmd+K) show staged local models,
in-flight downloads as live progress rows, and load-into-memory bars
- local-setup campaign tip for eligible hardware; System resources
statusbar widget (GPU/VRAM/RAM); in-chat load progress during sends
- friendly dead-server errors, and failed agent builds retry on the next
send instead of wedging the session
Co-developed with NVIDIA field feedback on RTX 5090 and DGX Spark.
The #94248 native half. A delegation deadline abandons the child's daemon
worker while it is typically parked inside an in-flight OpenSSL read
(Codex Responses stream / httpx). PR #90889's deferred close (cherry-picked
here, authorship preserved) stops the timeout thread from closing the child
under the running future — but the deferred close only fires once the worker
unwinds, and a worker blocked in ssl.read never unwinds on its own: the
cooperative interrupt cannot reach a thread inside OpenSSL, so the child's
SessionDB, httpx pools, and subprocesses stayed pinned until process exit,
and any path that still hard-closed the transport released FDs under a live
SSL BIO (the #29507/#67142/#70773 native-corruption family; SIGSEGV 17-72ms
after "Subagent N timed out" on macOS arm64).
Fix — bounded drain after deferral:
- AIAgent._drain_transports_after_abandonment(): shutdown()-only sweep of
the shared client's pooled sockets (force_close_tcp_sockets — FD release
stays with the owning worker), abort+poison of the cached per-request
openai/anthropic wire clients, Codex app-server request_interrupt(), and
the inline _active_request_abort hook. Never client.close(), never
socket.close().
- delegate timeout path: after registering the deferred-close callback,
run one immediate drain plus one 5s re-sweep (covers a connection opened
between the interrupt and the first sweep). The settled read (EOF/EPIPE)
lets the worker unwind, which triggers the deferred close on the worker's
own thread — the only safe FD-release boundary. A worker that still never
settles retains its resources rather than risking a cross-thread close.
Live repro (Linux, real TLS server subprocess + real httpx client blocked
in OpenSSL read at the deadline + real SessionDB): before — child.close()
ran on the timeout thread with in_flight_ssl_read=True (client FDs released
under the live read; #94736 self-heal WARNING fired on the worker's unwind
flush); after — drain settles the read in ~1ms, worker unwinds, close runs
on the worker thread with in_flight_ssl_read=False.
Not live-tested on macOS arm64 (no macOS runner); the fix is
platform-neutral teardown ordering proven on Linux.
Closes#94248
A gateway process opened state.db from ~12 call sites, each minting its
own writer connection, self._lock, close-time WAL checkpoint, and
token-writer thread. With N independent writers on one WAL file, one
connection's close-time checkpoint could race another's growth — the
lost/reordered-page-write signature across 11+ incidents (#90837).
Adds hermes_state_registry.py: a process-wide, per-path, refcounted
shared registry owning the writer boundary.
- acquire(path): same resolved path returns the same instance (one
writer connection, one lock, one token-writer thread) for every
long-lived in-process caller (gateway runner, SessionStore, per-agent
lazy recall, cron per-job, mirror, channel_directory, slash_commands,
shutdown_flush, session_search, react_to_message, delegate, mcp_serve,
auto_archive, tui_gateway).
- close() on a shared instance is a NO-OP — the registry owns the
lifecycle, so one caller's close can never tear down a writer other
callers still hold.
- Generation-aware retirement on inode change: a replaced state.db
RETIRES the live generation (never lent again) but keeps it alive for
existing holders; release is object-keyed so holders of the old
generation drain it independently of the new one. The old
generation's own write path still fails with the typed
StateDbReplacedError (existing protection, unchanged).
- Replacement-open failure leaves NO registry entry for the path —
the next acquire retries fresh, never hands out a closed stale object.
- All teardown runs OUTSIDE the registry lock: a final release's WAL
checkpoint can never stall acquisition for every state.db.
- close_shared_session_dbs() at gateway shutdown drains every
generation (live + retired) as the final safety net.
CLI one-shots, recovery flows, and read-only cross-profile opens keep
using SessionDB() directly with their own close() — only long-lived
in-process sites route through the registry.
References #90837 (root-cause tracker stays open: the #10 EOF signature
and the WAL-lifecycle A/B verdict remain under investigation there).
The turn-lease timeout/interrupt paths return from inside the try block
before set_affinity_scope() runs; the finally then read an unassigned
local -> UnboundLocalError. This was the cause of the 4 red
cross-process lease tests on PR #97158's CI.
(cherry picked from commit 05a3c8a4caa59831b01d13f3950e61385aadbfa5)
Every conversation-affinity hint Hermes sends is derived from the PHYSICAL
session id: prompt_cache_key on both OpenAI-wire transports, OpenRouter's and
Nous Portal's sticky session_id, and xAI's x-grok-conv-id. A host that mints
one physical session per RESPONSE re-keys all four on every reply, so the
conversation never lands back on the routing bucket it just warmed (#96811).
Two hosts do exactly that. Hermes Studio's group chat mints
gc_run_<room>_<profile>_<name>_<uuid4hex> per reply and destroys it after,
and POST /v1/responses with client-managed history mints str(uuid4()) per
request — while parsing X-Hermes-Session-Key one screen earlier and handing
it to the agent.
Hermes must not infer the logical conversation from the id's syntax: that
rule merges independent client-supplied ids and Studio members truncated past
its 96-character boundary (the #79017 failure class). It does not have to.
gateway_session_key is already the "stable per-chat key" built by
gateway.session.build_session_key from that header, and branching
deliberately does not key off it. The affinity path simply never consulted it.
- agent/prompt_cache_scope.py: declared_conversation_scope() resolves the key
into gwk_<sha256[:24]> and outranks the lineage walk (it is stable across
rotation AND across per-response ids). Hashed because, unlike a session id,
the key embeds platform/chat/user identifiers and leaves the process
verbatim as a sticky id and as x-grok-conv-id.
- agent/portal_tags.py: a separate ambient scope for ROUTING, published only
when a host declared one. The providers read the attribution id when it is
unset, so delegate trees keep sharing their parent's sticky key and every
host that keeps one id per conversation is byte-identical to before.
- hermes_state.py: is_explicit_fork_child() — the public view of the marker
rules that keep /branch children, delegate subagents and tool children off
their parent's chat key. Background-review forks clone the live runtime, so
_persist_disabled excludes them for the same reason (#79161).
Refs #96570Fixes#96811
Closes the #99758 review P1 (andrexibiza): with a generation claim in
play, `interrupt()` called `_admit_hard_cancel()` BEFORE the claim was
validated at the final mutation edge, and the production
`CompressionCommitFence.cancel_before_commit()` irreversibly sets
`_cancelled = True` whenever no commit has started. So a watchdog abort
that ultimately DECLINED (real progress landed in the window, claim went
stale) had already killed the recovered turn's legitimate pending
compression commit — `begin_commit()` refuses a cancelled fence forever.
Generation authority covered interrupt publication but not the
compression-fence mutation that preceded it.
Split hard-cancel admission into two halves:
- `_wait_for_compression_commit()` runs pre-claim and is NON-mutating:
it only blocks when `commit_in_flight` is true (the started-commit
branch of the production fence waits for `finish_commit` without
cancelling), so the interrupt still publishes only after an in-flight
SessionDB mutation has finished — exactly as before.
- `_cancel_pending_compression_commit()` runs AFTER
`_consume_claim_and_publish_first_state()` survives, so the
destructive pending-commit cancellation can never outlive a stale
claim. If a commit crossed its boundary in between, it is no longer
fence-cancellable and completes on its own.
Regression coverage (both use the real `CompressionCommitFence`):
- `test_declined_abort_does_not_cancel_pending_compression_commit`:
parks the interrupt at the claim-reservation release, lands real
progress (G+1), lets the interrupt decline, then proves
`fence.begin_commit()` still admits. Red on the pre-fix tree
(mutation-checked: the fence was left cancelled).
- `test_declined_abort_parks_and_leaves_fence_operational`: the
in-flight-commit window variant — activity lands while the interrupt
waits on a started commit; the interrupt declines and a fresh
`begin_commit()` still admits afterwards.
- The round-6 witness (`...resumes_inside_interrupt_publication`) now
models an in-flight commit (`commit_in_flight = True`) so its park
point stays inside the pre-claim wait, matching the new admission
shape.
Also updates the `interrupt()` docstring for the deferred destructive
cancellation.
Closes the #95663 round-8 review blocker (false settlement before
commit veto): the pre-commit surface (`_surface_stall`) logged
"Force-aborting the turn and stopping lease renewal" and warned the
user "aborting it so the session can recover" BEFORE `_commit_abort`
could veto — so a turn that resumed during the warning window (or an
exceptional interrupt path that declines fail-closed) was reported as
force-aborted with lease stopped while it actually continued running.
- Split the surface: `_surface_stall` is now observational only ("no
progress for Ns; attempting recovery"), and the definitive
aborted/lease-stopped settlement moves to a new
`_surface_committed_abort` that runs only after `_commit_abort`
succeeds and the turn lease is deactivated.
- Rate-limit repeated pre-commit surfaces per observed generation: a
turn whose aborts keep declining no longer re-logs an ERROR and
re-warns the user every poll interval.
- Add the committed-path regression test
(`test_watchdog_publishes_definitive_settlement_only_after_commit`)
and extend the declined-path witness
(`...resumes_during_warning`) to assert no committed-abort or
definitive pre-commit claim appears when the abort is vetoed. Both
fail on the pre-fix tree (mutation-checked).
- Document the `_interrupt_turn` lease-loss asymmetry (fires
unconditionally, no generation claim — losing the lease means the
process no longer owns the session).
- Trim review-round archaeology from comments/docstrings (keep the
WHY, drop the round numbering), and drop the dead
`cancel_event` compat note from the test fence.
- Document `agent.turn_liveness` in the configuration guide.
On top of PR #95663 by Finn763 (cherry-picked with authorship
preserved).
Add a turn-liveness watchdog keyed to the agent activity clock: a turn
that stalls mid-flight while the durable lease keeps renewing is logged
loudly, surfaced to the UI, force-interrupted, and — when the hard
interrupt cannot unwind the wedge — lease renewal is stopped so
stale-turn cleanup can reclaim the session.
Race safety (rounds 3/4/6 of the #95663 review, all folded into this
squashed commit):
- AIAgent.interrupt(require_generation=G) re-validates the generation
claim at the last instant before the hammer; a stale claim abandons
the abort and the turn continues.
- The claim is reserved under the activity lock, invalidated by any real
progress in _touch_activity(), and consumed immediately before the
first observable interrupt publication; exceptional paths fail closed.
- Claim consumption and the first interrupt publication are atomic
inside one _liveness_activity_lock() critical section; unclaimed
interrupts publish lock-free so AIAgent stand-ins without the liveness
seam keep working (CI 33096454629 regression, fixed here).
Deterministic race regressions (written red-first) in
tests/run_agent/test_turn_liveness_watchdog.py cover the
post-revalidation window, the consume-to-publication window, the
exceptional path, and atomic claim consumption.
Round-7 rebuild: single squashed commit on current origin/main; the
former four-commit lineage (241f8e484..299122558 on merge base
6defe7eb6c) no longer exists, so no surviving commit carries a red
exact-object CI record, and no empty CI-trigger commit was added.
Detect same-inode cp via a generation stamp, halt FTS repair, and divert
unwritten transcripts to sessions/<id>.jsonl plus the gateway pending spool.
Co-authored-by: Cursor <cursoragent@cursor.com>
Salvaged from #98741, composed on top of the merged #98424 preflight
fail-closed boundary. A host-ceiling compression timeout is now a typed,
thread-safe outcome consumed by every automatic caller:
- conversation_compression.py: threading.local + per-agent lock timeout
state (mark/reset/read helpers) upgrading #98424's simple attribute
where overlapping automatic/manual compression entrypoints matter;
the _last_compression_timed_out attribute stays as compat mirror.
- conversation_loop.py: the mid-turn pre-API pass and the provider
overflow (413/400 context_length_exceeded) recovery path end the turn
with the typed compression_exhausted recovery contract instead of
re-sending the unchanged oversized request and re-entering compression
in the same turn.
- run_agent.py/turn_context.py: forwarder resets the typed state per
attempt; the #98424 turn-start check reads it through the typed helper.
Tests: thread-safety/atomicity of the state helpers, overflow-recovery
non-re-entry, and typed terminal result.
When the default/global state.db is corrupt at gateway startup,
SessionStore degrades (_db=None) and record_gateway_session_peer never
self-heals. Under multiplexed profile routes the AIAgent's lazy
_ensure_db_session was then the ONLY durable write for the session, and
it created the row identity-less (session_key/chat_id/chat_type/
thread_id/user_id/origin_json all NULL) — unrecoverable by
find_latest_gateway_session_for_peer, so Telegram chats forgot prior
turns.
Persist the routing identity the agent already carries into
create_session; plain CLI sessions keep the old identity-less shape.
Salvaged leg 1 of #88804; the transcript-recovery leg is covered by the
scope-aware session DB resolution already on main.
Sessions created on the launch/default profile were persisted with
profile_name = NULL by all three writers (run_agent._ensure_db_session
None'd out 'default'; the desktop backend's _ensure_session_db_row and
session.branch passed None when no profile_home override was set).
NULL used to mean 'launch profile' by convention, but the desktop now
keys sessions by (profile, id), filters the sidebar by profile scope,
and resolves @session:<profile>/<id> deep links by profile match — a
NULL row matches nothing, so sessions created around a profile switch
vanished from the sidebar and their deep links could not be opened
(#99222). The #94724 one-shot legacy-owner backfill stamps literal
'default' onto old NULL rows, so writers minting NEW NULL rows after
that backfill ran recreated the exact state it exists to repair.
Stamp the real profile name at creation time in all three writers.
E2E-verified against a temp HERMES_HOME: both the desktop create path
and the agent path now persist profile_name='default'.
Fixes#99222
record_timeout_failure() now persists
'backoff:<failure_kind>:strategy=<tail_mode>' into the state.db
cooldown row (sessions.compression_failure_cooldown_until +
compression_failure_error), so a failed/stalled/cancelled attempt's
identity survives gateway restarts and the rebuilt compressor makes the
same skip decision via bind_session_state()/get_active_compression_
failure_cooldown(refresh=True). Host callers pass ceiling_exhausted /
stalled; the stall-interrupt path passes stall_interrupted. A
successful compression still clears the row.
Stage destination native-compaction capabilities until the complete runtime and context setup succeeds, and restore them with primary and fallback runtimes. Keep native compaction default-deny across live switches and session reconstruction.\n\nVerification: uv run --with pytest --with pyyaml python -m pytest tests/run_agent/test_switch_model_context.py tests/run_agent/test_native_compaction.py tests/run_agent/test_native_compaction_switch_capabilities.py tests/run_agent/test_switch_model_rollback.py tests/run_agent/test_fallback_reasoning_override.py tests/run_agent/test_primary_runtime_restore.py tests/run_agent/test_provider_fallback.py -q -o 'addopts='; uv run --with ruff ruff check <touched files>; git diff --check
Salvaged from PR #97815 by @itsflownium, slimmed to the schema-free core:
- TodoStore gains a monotonic in-memory revision; the todo tool result
returns it so clients can reject stale updates
- tui_gateway emits a dedicated todo.updated full-snapshot event that
bypasses optional tool-progress display settings
- session resume/activate responses attach the authoritative todo
snapshot; renderer restores it with revision arbitration
- desktop store tracks per-session revisions and rejects regressions
The session_todo_state DB table from the original PR is intentionally
dropped: canonical todo tool results already persist in conversation
history, so resume paths derive the snapshot from the stored transcript
instead of a parallel store.
- Preserve streamed assistant text in Desktop UI when message.complete delivers empty text.
- Prevent destructive hydration in Desktop useMessageStream over rendered text on empty completion.
- Recover stream buffer in finalize_turn when final_response is empty on healthy turns.
- Unify in-place blank assistant repair, watermark clone resolution, non-blank concurrent winner adoption, and batch row appends into a single atomic guarded SessionDB transaction.
- Synchronize canonical committed content to live in-memory messages dicts and preserve all-or-nothing rollback semantics on persistence failure.
Every provider response carries usage.prompt_tokens — exact ground truth
for the full request (system prompt + tool schemas + history). Context-size
checks now anchor on the last main-loop response's usage and estimate only
the messages appended since, instead of re-estimating the whole history
with chars/4 heuristics and flat 1500-token image costs. The estimate error
window shrinks from the entire conversation to one turn and self-corrects
at every response.
- agent/model_metadata.py: capture_usage_anchor() / anchored_context_tokens()
with a structural base-message identity check that fails closed on any
transcript rewrite.
- agent/conversation_loop.py: anchor captured at the single main-loop usage
site (MoA uses pre-fold aggregator usage; advisor/aux calls never anchor);
pre-API pressure check prefers the anchor.
- agent/turn_context.py: preflight compression estimate prefers the anchor.
- agent/context_breakdown.py: /context display prefers the anchor.
- Invalidation: compaction rewrite (conversation_compression), codex native
compaction (codex_runtime), session reset/switch (run_agent), plus the
fail-closed structural check for splices/micro-compaction.
- Usage-less responses keep the previous anchor; no anchor -> pure
estimation fallback (first request of a session).
Follow-up to the #96217 salvage: the codex/xai/github route checks were
re-implemented inline at four sites (codex_responses_adapter helpers,
chat_completion_helpers kwargs build, _is_openai_codex_backend, the
run_agent silent-reject hint). Consolidate them into
classify_responses_route() / ResponsesRouteFlags in
codex_responses_adapter and migrate every site — backend-identity
predicate class (#22548/#70893/#59561/#72468).
Host checks use exact-host-or-subdomain semantics, never substring
matching.
When context-compression rotation fires mid-turn, the current user
message was persisted twice into the child session. Root cause: dedup
used id()-seeded sets of copies instead of markers on the live objects.
Replace with _DB_PERSISTED_MARKER-based dedup as the sole authority:
- _ensure_compressed_has_user_turn returns CompressedUserTurnOutcome
- After publish_compression_child succeeds, stamp the live anchor-source
row (not a drifted index) with _DB_PERSISTED_MARKER
- _sync_persisted_markers mirrors stamps from result to live lists by
scoped identity (handles direct-path, adoption divergence, _session_messages)
- Remove _flushed_db_message_ids from rotation commit path (markers replace it)
- Unconditional (loud) imports — no silent fallback
Salvage of #94996 by @fedosis, rebased on top of #95433 (stall-fallback,
already merged). Both conversation_compression.py and run_agent.py are
built from origin/main + #94996's diff applied on top, preserving the
force_terminal refactor and _publish_new_fence from #95433.
Credit: @fedosis original PR #94996.
A stalled compression summary never raises, so the auxiliary client's
exception-path fallback is unreachable from it. When the progress-aware
timeout aborts a stalled worker, re-run the summary once pinned to the
first auxiliary.compression.fallback_chain entry before degrading to
continue-without-compression.
The pin is a single-use ContextVar consumed by the context compressor's
summary call, so it cannot leak into the detached stalled worker or the
compressor's own main-model retry. A fresh fence is minted through the
host factory so a /stop during the retry still admits against the live
commit boundary.
* refactor(prompt): remove the Nous Subscription block from the system prompt (~1.2K tokens/call)
* chore: retrigger CI (zero-job dispatch failure, auto-heal)
