The native Responses stream does carry summary_index, so the part boundary is
structured data here rather than something to infer. Break on a change of
index, and leave streams that send no index (plain reasoning_text) untouched.
- Single source for the approval-derived bound: public human_wait_ceiling()
in tools/approval.py; the gate's lock-timeout helper delegates to it
instead of re-deriving timeout + margin (was duplicated in two modules
and reached for a private _get_approval_timeout).
- Shared _clamped_window_seconds() for the close-time accrual and the
open-window read, so the two clamps are identical by construction.
- Gate __init__ grows session_key kwarg; tests construct via the real
constructor instead of mutating privates post-hoc.
- Gateway test resolves its pending approval via resolve_gateway_approval()
(the production /deny path) instead of hand-rolling queue-entry internals.
- Docstring accuracy: human_wait_seconds monotonicity caveat under cap
eviction; s/pre_tool_block/pre_tool_call/ hook name.
Review-driven follow-up to the #79719 fix:
- Clamp the CLOSE-side accrual too: a wedged window that eventually closed
used to inject its full unclamped overstay into completed_seconds,
retroactively extending a running batch's deadline by hours. Both clamps
now share one ceiling helper (_human_wait_ceiling = approvals.timeout +
HUMAN_WAIT_MARGIN_S), and the gate's lock-timeout uses the same margin
constant so the bounds cannot drift apart.
- Evict idle sessions until the table is under the cap (was: at most one
per insert, so churn could outgrow _HUMAN_WAIT_MAX_SESSIONS). Entries
with an open window are still never evicted.
- Log (debug) instead of silently swallowing a failed session-key snapshot
in the gate constructor.
Tests: close-side clamp regression + table-cap assertion added; suite at
17 passed.
A tool wedged inside _ConcurrentToolAuthorizationGate hung the whole turn
forever (#79719): excluded_seconds() measured residency in gate.run() —
arbitrary code — so an open window grew 1:1 with wall clock and the batch
deadline's remaining was constant (remaining = deadline - window_started;
now cancels out). A hanging pre_tool_call plugin or an approval round-trip
to a dead client defeated the deadline entirely. The serialization lock was
also an unbounded acquire, so every other worker needing authorization
parked behind the wedged holder forever.
Fix, in two halves:
- tools/approval.py grows per-session human-wait accounting
(human_wait_window / human_wait_seconds). The two places that are
verifiably blocked on a HUMAN — the CLI approval prompt and the gateway
approval poll loop — mark their own windows. Both are intrinsically
bounded by approvals.timeout; the open-window read is additionally
clamped to that timeout plus a margin as belt-and-braces.
- _ConcurrentToolAuthorizationGate keeps only serialization, with a bounded
acquire (approvals.timeout + 60s; on expiry the prompt runs unserialized —
the same degradation the start-order gate accepted in #79705).
excluded_seconds() becomes a baseline-delta read of the session's
human-wait total.
A wedged plugin now contributes nothing to the exclusion, so the batch
times out at the normal deadline with correctly labeled results, while a
genuine approval wait — which can legitimately exceed any fixed bound —
still extends the deadline in full. E2E (real AIAgent, worktree imports):
wedged-plugin batch on main never ends (>30s observed, 3s deadline); with
the fix it ends at 3.0s. A 4s simulated approval over a 2s deadline
completes without a timeout label.
Closes#79719
Follow-up to the salvaged start-order gate bound. Two gaps remained, both
reachable through the same knob.
1. The gate bound ignored the batch deadline it sits under. With
HERMES_CONCURRENT_TOOL_TIMEOUT_S below 120s the deadline fired first, so
the parked tools were still reported as "timed out" without ever running --
the exact bug the bound exists to fix. The gate now clamps to
min(120s, batch_timeout / 2), matching the sibling constant's documented
habit of relating the two timeouts.
2. A gate-parked worker released purely by its own timeout could wake up after
the batch was abandoned and dispatch its tool anyway: wasted work whose
result nobody reads, a duplicate post_tool_call for a tool_call_id the turn
already closed as timeout, and agent._current_tool left pointing at a dead
tool for the rest of the session (the main thread's reset already ran).
Abandonment is now a first-class wakeup: both abandon sites set an event and
notify the condition, and a released worker raises _BatchAbandoned instead
of dispatching. Parked threads are reclaimed in milliseconds rather than one
full gate timeout plus a tool runtime.
Also names the tool in the gate-timeout warning. The closure's function_name
binds the last-parsed tool, so logging it directly would have printed the wrong
name; it is threaded through _begin_in_order instead.
Measured, 3-tool batch with the first tool wedged during dispatch:
main PR as-is with this commit
dispatched in batch 0 0 tool_b, tool_c
dispatched after return 0 2 (ghost) 0
_current_tool leaked no "tool_b" no
Adds tests/run_agent/test_start_order_gate.py (3 tests). Mutation-checked
against the parent commit: the starvation guard passes there (it binds the
salvaged fix), while the deadline-clamp and abandonment guards both fail,
reproducing the ghost dispatch as
"tool(s) dispatched after the batch was abandoned: [tool_a, tool_b]".
* feat(agent): read_preview — the desktop-gated tool that reads the in-app browser
The agent could open the preview pane (open_preview) and read the embedded
terminal (read_terminal), but the browser it had just opened was a black box —
'what does this page say?' had no answer. read_preview mirrors read_terminal
end to end: HERMES_DESKTOP-gated via check_fn (zero schema footprint outside
the GUI), dispatched through the same agent callback pattern, windowed with
start/count so a long page pages instead of flooding context.
* feat(gateway): preview.read blocking bridge
Same lifecycle as terminal.read: the tool blocks on preview.read.request, the
renderer answers preview.read.respond (allow_expired — a slow page extraction
losing the 45s race must not surface a raw 4009), and a timeout emits
preview.read.expire so late answers resolve quietly.
* feat(desktop): the renderer serializes the active preview tab for the agent
preview-reader.ts is the preview analog of the terminal's buffer registry: the
URL pane registers a page reader (webview executeJavaScript → title + visible
innerText) keyed by tab id; readActivePreview resolves the ACTIVE tab, windows
the text (24k cap per read), and answers file/artifact tabs with identity plus
a note pointing at the tool that reads that content directly. The gateway
event handler answers preview.read.request beside terminal.read.request.
Three fakes pin mark_exhausted_and_rotate's signature explicitly and broke on
the new argument. They now assert it rather than just tolerate it — the xAI
spending-limit case is exactly the billing-403 this fixes, so it should be
pinning `failure_reason == "billing"`.
* fix(agent): adopt .env credential/base-url edits at the turn boundary
A Settings save (desktop PUT /api/env, hermes setup) updates .env and
the saving process's os.environ, but a live session worker keeps the
base_url/api_key captured at agent init until restart — an open chat
silently kept calling the old endpoint (e.g. a local-server key sent to
api.openai.com, failing with an opaque 401).
Add AIAgent._try_refresh_env_client_credentials(), called at the start
of each conversation turn: re-resolve the provider's env-sourced
credentials (load_env() is mtime-memoized, so an unchanged file costs
one stat()) and rebuild the client via the existing
_replace_primary_openai_client machinery when the user edited them.
The refresh reacts only to env edits — resolved values changed since
the last look — never to mere divergence from the agent's current
values: credential-pool rotation and failover legitimately move the
session off the env credential, and stomping those back would flap.
Config model.base_url / pool custom endpoints keep precedence: edits
are only adopted while the session still runs on the registry default
or the previously-seen env value.
Lift _get_env_prefer_dotenv out of _seed_from_env to module level
(get_env_prefer_dotenv) so both the pool seeder and the per-turn
refresh share the same .env-over-os.environ resolution, including the
op:// indirection handling.
Fixes#67821
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(agent): address sweeper review on env credential refresh
- Cover named custom providers (#67935): provider="custom" has no
PROVIDER_REGISTRY entry, so resolve the config block's key_env through
the same lookup the runtime resolver uses.
- Make the edit baseline transactional: a failed client rebuild rolls the
agent back and leaves _env_creds_seen un-advanced so the unchanged edit
is retried next turn.
- Recompute route-derived TLS material and default headers on a base-url
change, via a _reapply_route_client_config helper shared with
credential-pool rotation so the two paths cannot drift.
- Rebase onto main: get_env_prefer_dotenv keeps the scoped _get_secret
semantics from the profile-isolation fix (no raw os.environ reads).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: map jskang@lablup.com to rapsealk
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Brooklyn Nicholson <brooklyn.bb.nicholson@gmail.com>
The output-cap retry loop reduced max_tokens by 64 tokens per attempt but
never called _compress_context(), so the compressor never fired. Input
growth (~65 tokens/attempt) canceled the savings, leaving the session
stuck at 200,001 tokens — 1 over the 200,000 ceiling.
The fix adds compression to the output-cap retry path. The compressor
drops the middle window, freeing ~50% of tokens. If compression makes
>=5% savings, the session continues; otherwise vision payloads are
stripped or the session ends with compression_exhausted=True.
Also adds CHANGELOG.md entry and bug fix report.
Local tool-call templates can emit a bare bracketed token (e.g. "[memory]")
as assistant content alongside a function call. The loop treated that
protocol scaffolding as visible content: it got cached as the post-tool
fallback, and when the next turn came back empty, the marker was replayed
as the final response and written into the persisted transcript. Later
context compaction preserved that history, letting the model repeat the
marker in subsequent turns.
Detect content that is only a bracketed marker (`[name]`) when the
response also carries tool_calls, and drop it before it can be cached
or persisted. Scoped narrowly: only fires alongside tool_calls, so a
genuine final response of "[memory]" without a tool call is unaffected.
Salvage of #65260's b7d7cfd0e (ported — the PR's close() predates ~4K
commits of teardown-step churn, so the hunk is re-anchored after step
6b rather than cherry-picked).
agent/codex_runtime.py already drops _codex_session on turn crash and
on retirement, but AIAgent.close() — the hard teardown for /new,
/reset, and session expiry — had no owner for it, so the app-server
child process survived until interpreter exit. Long-lived gateways
accumulate one leaked subprocess per ended Codex session.
The attribute is cleared BEFORE close() so a concurrent reader can't
observe a half-closed session and a raising close() can't strand a
stale reference (tested).
Tests extend the author's original lifecycle test with the
raising-close and no-codex-session cases.
Review follow-up on the #30223 salvage: the original changed the base
cooldown from 60s to 1800s, benching the primary for 30 minutes on the
FIRST 429 (30x regression in primary-restore latency) and breaking the
existing test_rate_limit_exhaustion_keeps_60s_cooldown contract.
Keep upstream's 60s base and escalate per consecutive rate-limit:
60s -> 2m -> 4m -> 8m -> ... capped at 4h. Counter still resets on
successful primary restore (cicae's mechanism, unchanged).
New tests: escalation doubling, 14400s cap, reset-on-restore.
Existing 60s contract test passes UNCHANGED. Mutation-checked:
escalation disabled -> 2 fail; reset disabled -> 1 fails.
The flush now goes through append_messages_batch; MagicMock-based
assertions and barrier fakes that hooked append_message observed
nothing (the flush's try/except swallowed the AttributeError). Assert
on the batch payload instead.
Re-derivation of #23254 (@devsart95) on today's flush loop. The turn
flush in _flush_messages_to_session_db wrote one BEGIN IMMEDIATE
transaction per message row; a typical agent turn (user + assistant +
tool results) paid 3-8 transactions -- and, off WAL (the default on
macOS while the WAL-reset guard is active), 3-8 fsyncs -- per turn.
Adds SessionDB.append_messages_batch: same row shape as append_message
(shared _prepare_message_row serializer + _MESSAGE_INSERT_SQL column
list, so the two writers cannot drift), same compression-lock and
compression-closed guards, one aggregated session-counter UPDATE, one
transaction for the whole batch. Row serialization stays outside the
write lock.
The flush loop now collects the turn's new rows and writes them in one
call. All-or-nothing pairs exactly with the persisted-marker stamping:
on failure no rows landed and no markers were stamped, so the next
flush re-writes the whole tail (same recovery contract as before,
minus the partial-prefix case that could double-count).
Measured (same harness, 5-message turn, journal_mode=DELETE,
synchronous=FULL): 2.32ms -> 0.83ms median per turn flush (64% faster,
5 fsyncs -> 1). On WAL the win is smaller but the atomicity fix holds.
Cross-PR interaction fix: #77714 (salvage of #71775) changed
_available_entries to return (available, pending_refresh) while #77631
(salvage of #67642) added next_available_at() which still truthiness-
tests the bare return. A non-empty tuple is always truthy — even
([], []) — so the reset-aware gate silently returned None ('no wait
info') for every exhausted pool, disabling the feature #77631 shipped.
Unpack the tuple and test the available list.
Also adapts the lock-probe test for the RLock introduced by #77714
(same-thread non-blocking acquire always succeeds on an RLock; probe
from a helper thread instead).
Review fold on the #67642 salvage: next_available_at() called
_available_entries() — which prunes DEAD entries, syncs tokens, and
persists — and iterated self._entries with no lock, racing concurrent
select()/rotation exactly as has_available()'s comment warns. Wrap the
method body in self._lock and pin it with a non-blocking-acquire probe
test.
restore_primary_runtime retries the primary every turn once the 60s
transient cooldown clears. For subscription-window limits (Claude
Pro/Max 5h windows, Codex weekly caps) the reset is hours or days away,
so every retry is a guaranteed failure costing two provider switches
and two prompt-cache invalidations per turn.
Add CredentialPool.next_available_at() (earliest reset across exhausted
entries; None when available now or no reset info) and gate the restore
on it: skip while the primary's pool says nobody can serve, restore on
the first turn after the reset elapses. Fail-open: any gate error or
missing reset info falls through to the existing per-turn retry, so
recovery can never be later than today. Cross-provider fallbacks
consult the PRIMARY's pool (not the attached fallback pool), reusing
the loaded pool for the existing rebind to keep auth reads at one per
restore.
The retry loop gates on real time.time() < sleep_end; with sleep mocked
to a no-op the test hot-spun 7.5 wall-clock seconds. Advance a fake
clock by each sleep amount instead (pattern precedent:
test_session_activity_persist.py).
Empty content retries previously fired back-to-back with no delay,
wasting up to 3 rapid API calls, and could not be cancelled mid-wait.
Apply the same jittered_backoff() already used for rate-limit and
API-error retries, sleeping in small increments so a user interrupt
aborts the wait instead of blocking until it elapses.
Fixes#35230
OpenCode Zen's relay rejects the Anthropic-style content block format
that cache markers produce (content becomes a block array instead of a
plain string), causing HTTP 400 with "content must be string, not block
array" for DeepSeek models.
Reverts the DeepSeek addition from commit 6b6435a874 while preserving
the Qwen/Alibaba caching path which continues to work.
Fixes#77217
Post-tool compression path passed context_compressor.last_prompt_tokens (0 in the no-usage
fallback) to _compress_context instead of the overhead-aware _real_tokens computed just above
— same tool-blind bug as the overflow handlers (upstream PR #77169 review, teknium1). Also adds
production-path regression tests asserting the 413, context-overflow (two wordings), and
Anthropic long-context recovery handlers pass estimate_request_tokens_rough(..., tools=...)
(sentinel-patched) to _compress_context.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
strip_think_blocks passed the same response-scrubbing strings through
re's pattern dispatcher on every response. Skills Guard repeated the
same work for 121 patterns against every scanned line.
Compile the existing expressions once and reuse Pattern.sub/search. Keep
each generic tool-call tag in its own paired expression so mismatched
openers retain their payload while existing stray-closer cleanup remains
unchanged.
Part of #33208
Salvaged from #32713 by @ErnestHysa.
Co-authored-by: ErnestHysa <takis312@hotmail.com>
Heartbeat write discipline for the durable SessionDB activity projection:
- Pin the cadence in a named constant
(SESSION_ACTIVITY_HEARTBEAT_MIN_INTERVAL_SECONDS = 60s, contract >= 30s,
deliberately config-independent so no compression.*/agent.* setting can
turn the heartbeat into a high-frequency writer on the contended
SessionDB write path).
- The write already rides the standard _execute_write patience path via
SessionDB.touch_session_activity — verified, now documented in the
docstring.
- Best-effort hardening: a failed heartbeat write never raises into the
agent loop; the bare 'pass' becomes an explicit debug log with traceback.
- Tests: direct proof that a heartbeat DB failure doesn't propagate,
the cadence constant is pinned >= 30s, and the rate limiter keys off
the shared constant (boundary tested on both sides of the window).
Three mechanisms to detect and notify when gateway sessions stall silently:
1. Mid-turn activity heartbeats stamped to SessionDB so hermes sessions list
and hermes status show progress during long turns without new message rows.
2. Stall watchdog: when a busy session has pending inbound and the shared
activity clock is idle past agent.session_stall_timeout (default 300),
log a WARNING and notify the user once to try /new. Notify-only; does
not kill the turn.
3. Compaction timeout: fenceless compress_context callers get a progress-aware
host budget (compression.context_timeout_seconds default 120 idle,
compression.context_total_ceiling_seconds default 600 ceiling). On timeout,
cancel via commit fence, skip compaction without dropping messages, and
continue the turn.
Closes#72016 (slices 1-3; slice 4 cumulative SSE stream-retry deadline
remains a follow-up).
Cherry-picked from PR #72424 by @fangliquanflq.
Completes the cache-parity bug class from #76938: the parent's request
body diverges from the fork's not only at the ephemeral system prompt
but also at prefill messages (inserted right after the system message
at API-call time) and, on OpenRouter, at upstream-provider selection
(prompt caches live per upstream; an unpinned fork can be routed to a
different upstream and miss a byte-identical prefix).
Also hardens the tests: pairwise asserts instead of re-implementing the
production prompt join, and routed-path omission guards for the whole
gated kwarg family.
Makes interrupt-protected context compression cancellable by an explicit
user or lifecycle stop, without weakening protection against ordinary
incoming messages, voice interjections, or active-turn redirects.
Separates explicit hard cancellation from ordinary interrupt/redirect
state with a dedicated threading.Event; introduces
AuxiliaryExplicitCancellation as an attempt-local frozen-cause signal;
isolates the synchronous provider callback in a bounded daemon worker
during protected compression; atomically linearizes Codex timeout
cleanup against explicit cancellation; propagates hard cancellation
through child agents and explicit stop surfaces; serializes hard-cancel
admission against compression commit admission with
CompressionCommitFence; aborts before session rotation or late DB commit,
restores in-place transcript mutations and compressor state, and releases
the heartbeat and compression lease.
Based on #74449 by @suparious. Resolved merge conflicts in
agent/context_compressor.py (feasibility check + try/except) and
tui_gateway/methods_session.py.
The terminal-command guard path loaded config 2-3x per invocation via
load_config(), which pays a defensive deepcopy of the entire config on
every call (~356us of the ~376us warm-cache cost measured on a real
config.yaml). All six swapped call sites were audited read-only — every
caller takes scalar reads or iterates the returned structures; none
mutate (the save path at save_permanent_allowlist keeps load_config) —
so they now use load_config_readonly(), the API built for exactly this
(precedent: #74211, #74322; the one unsafe-site lesson from #56085's
salvage is covered by the mutation audit and a cache-integrity test).
Measured (real config.yaml, warm cache): load_config 376.0us ->
load_config_readonly 19.9us (18.9x); full guard pass
check_all_command_guards('ls -la','local') 930.7us -> 241.8us (3.85x).
Tests: new test_approval_config_readonly.py drives the real functions
against a temp HERMES_HOME — readonly call counts per function, a
no-deepcopy pin for the full guard pass, and cache-identity/integrity
checks. Existing test mocks retargeted from load_config to
load_config_readonly (same injection intent). Note: 6
test_approval_mode_parity failures are pre-existing ordering flakes —
identical with the change stashed on clean main.
The sequential executor's KeyboardInterrupt handlers emitted a cancelled
post-tool-call event for the current tool, called agent.interrupt(), then
re-raised — WITHOUT appending a tool result message for the interrupted call
or any remaining calls in the batch. The assistant tool-call turn was left
with no matching tool results, a message-role alternation violation that
malforms the next provider request (relying on downstream repair passes to
patch it, which don't run on every path).
The cooperative-interrupt block (_interrupt_requested) and the concurrent
executor already emit a result for every call_id; this brings the two hard-
interrupt handlers into line via a shared _append_cancelled_tool_results
helper that appends a cancelled result for the current + remaining calls
before re-raising.
Verified live before/after (0 tool results -> 3 for a 3-call batch
interrupted on the first tool) and with a sabotage-checked regression test.
52 interrupt/executor tests pass.
The parallel tool-batch planner treated search_files as unconditionally
parallel-safe (_PARALLEL_SAFE_TOOLS) with no path reservation, so a
batch of patch(path=X) + search_files(path=dir(X)) landed in one
concurrent segment and the search could observe pre-mutation file
content — a same-block write->read stale-read race.
Fix the class, not the site: path-scoped reservations now carry a
reader/writer role.
- search_files joins _PATH_SCOPED_TOOLS as a READER, reserving its
search root (default '.', matching the tool's default) instead of
bypassing path checks entirely.
- Overlap only conflicts when a WRITER is on either side: a write into
a searched/read subtree splits segments (ordered behind the write),
while reader<->reader overlap — previously split needlessly — now
stays parallel (concurrent reads commute).
- write_file/patch keep their existing writer barrier semantics.
Prior art surveyed for this design: Codex CLI's RwLock read/write
barrier (readers share, writers exclusive), Claude Code's
isConcurrencySafe partitioning, and gemini-cli's contiguous
parallelizable batching — all converge on reader-shared/writer-
exclusive with contiguous-order preservation, which this planner
already had for read_file/write_file/patch; this closes the
search_files gap and adds the missing reader/reader concession.
Verified by sabotage run (tests fail against the old planner) and an
E2E script exercising the real planner + real file I/O.
Setting prompt_caching.cache_ttl to a falsy value (false, null, off,
disabled, no, none) now fully disables prompt caching instead of
being silently ignored.
The disable propagates through anthropic_prompt_cache_policy() (early
return when _cache_disabled flag is set) and restore_primary_runtime()
(override after snapshot restore), so it survives /model switches and
fallback re-derivation — the gap that caused #56105 to be reverted in
#56126.
Salvage of #33555 by @BB-light, with model-switch/fallback survival
gap fixed on top.
Co-authored-by: BB-light <BB-light@users.noreply.github.com>
The bounded flush-scan in _flush_messages_to_session_db_unlocked skips
the identity-matched prefix of its previous snapshot, on the documented
assumption that no code path pops _DB_PERSISTED_MARKER from a live dict
in place. finalize_turn's pure-tool-call-tail fill is exactly that path:
it pops the marker so the filled content gets re-persisted — but the
cursor then skips the row anyway, so the delivered final response never
reaches state.db and /resume replays content="" (the #43849/#44100
class resurfacing via the perf cursor). Invalidate the cursor at the
pop site so the filled row is re-examined.
brooklyn!
kshitij
Ryder Freeman
Jeffrey Quesnelle
burak33bb
Alex Fournier
Jeongseok Kang
BobClawblaw
Hermes Agent
joaomarcos
JR Razmus
kshitij
EndeavorYen
devsart95
WojtekMR3
arimu1
JonthanaHanh
LFDM Core
Eugeniusz Gilewski
Teknium
fangliquanflq
fangliquanflq
Quark Assistant
Shaun Prince
spfcraze
Fangliquan
Rod Boev
BB-light
Xipong
wz-heng
Zeraphim