Follow-up to the cherry-picked #68258 base: the cross-session-stable
prefix (_cached_system_prompt_static) was only recorded on fresh
builds, so two paths silently degraded to the legacy single-breakpoint
layout (flagged in review of #68258/#69341/#69704):
- Session restore: gateway surfaces build a fresh AIAgent per turn and
restore the persisted prompt verbatim from the session DB; the static
prefix stayed None from turn 2 onward, flip-flopping the wire layout.
- Post-compression cached-prompt reuse: _invalidate_system_prompt()
clears the static prefix, and the keep-cached-prompt branch never
restored it.
Both sites now reconstruct the stable tier and adopt it ONLY when the
authoritative prompt string literally startswith() it — stable-tier
drift (skills edited, identity changed) falls back to the legacy layout
with the stored bytes untouched. Fail-open on any builder error. The
restore-path rebuild is gated on _use_prompt_caching so non-Anthropic
routes skip it entirely.
Refs #68191
Co-authored-by: JonthanaHanh <92574114+JonthanaHanh@users.noreply.github.com>
Co-authored-by: joaomarcos <joaomarcosdias444@gmail.com>
Co-authored-by: webtecnica <webtecnica@users.noreply.github.com>
The conversation loop normalizes message text right before the API call so
the request prefix is byte-identical across turns -- the stated reason is
KV cache reuse on local inference servers and better cache hit rates on
cloud providers. Cache breakpoints were injected *before* that pass, which
defeats it.
`_apply_cache_marker` rewrites a plain-string `content` into a
`[{"type": "text", ...}]` block. The normalization pass is guarded on
`isinstance(content, str)`, so every message that just got marked is
silently skipped by it and keeps its raw leading/trailing whitespace. A
message is only marked while it sits in the last-3 window, so:
turn N in the window -> marked, content "file1\nfile2\n"
turn N+1 rolled out -> plain, content "file1\nfile2"
The same logical message is sent with different bytes on consecutive
turns. The prefix stops matching at that position -- which is inside the
span the breakpoints were placed to protect -- so the reusable prefix
collapses back toward the system breakpoint on every turn. Tool results
carry a trailing newline almost by default (any shell command output), so
this is the common case, not an edge case.
Move the injection below every message mutation. Besides fixing the
whitespace divergence this stops breakpoints from being spent on messages
that the orphan sweep or the thinking-only drop is about to remove or
merge away -- a marker on a dropped message is a wasted breakpoint out of
the four available.
Nothing between the old and new call sites reads `cache_control`, and the
mutators now see the plain-string shapes they were written against.
Review follow-up for the dropped tool-call recovery (#69630): the
re-prompt pair was tagged _dropped_toolcall_nudge, but that marker was
not part of the ephemeral-scaffolding contract. _persist_session /
_flush_messages_to_session_db would therefore write the synthetic
'issue the actual tool call now' user message (and the narration-only
interim assistant turn) as real transcript rows — a resumed session
could replay the internal retry instruction as user-authored context
and prompt unsolicited tool use.
- Add _dropped_toolcall_nudge to _EPHEMERAL_SCAFFOLDING_FLAGS
(run_agent.py) so both SQLite and JSON persistence skip the pair.
- Add it to _SYNTHETIC_USER_FLAGS (conversation_compression.py) so the
compressor never treats the nudge as human intent.
- Flag the interim assistant half of the pair too, and include the
marker in the finalization scaffolding pop so a genuine turn end
strips the pair from the live transcript (mirrors the
_empty_recovery_synthetic pattern).
- Regression tests: flagged messages classify as ephemeral, the
returned transcript contains no scaffolding, and the turn tail stays
on the real assistant answer.
The initial fix guarded the no-tool-calls else branch, but that branch only
SETS final_response — the turn actually finalizes later, in a separate block
after final_msg is built. Runs that reached finalization via that path exited
without the guard ever running (observed live: a scheduled PR reviewer stalled
at tool_turns=1-2 with zero recovery nudges).
Move the recovery to the finalization chokepoint (right after final_msg is
built), so it catches every path that ends a turn. Single guard now:
- increments a consecutive-stall counter and re-prompts (bounded to 3),
- resets on any successful tool round, and
- resets on a genuine (non-mismatch) turn end,
so it guards each stall independently without capping the whole run and
without looping forever.
Verified live: the scheduled reviewer now recovers through the stalls and
submits real reviews — PR 57800 APPROVED, PR 54826 COMMENTED — one PR per run.
Some providers (observed: claude-opus-4.8 / claude-sonnet-4.5 on GitHub
Copilot, ~2026-07) return finish_reason="tool_calls" while the parsed
tool_calls array is empty — the model signalled it wanted to act but the
payload shipped no call. The conversation loop took the no-tool-calls
else branch, treated the turn's narration as the final answer, and exited
with the task unstarted.
On unattended multi-step jobs this is silent failure: a scheduled PR
reviewer, for example, would narrate "Let me verify the PR..." and stop
at tool_turns=0 every run, never submitting a review, while the job still
reported success.
Fix: in the no-tool-calls branch, detect the provider contract violation
(finish_reason == "tool_calls" with zero tool_calls) and re-prompt the
model to emit the call instead of exiting. Bounded to 3 consecutive
stalls; the budget resets after any successful tool round so it guards
each stall independently rather than capping the whole run. The narration
may live in content or only in the reasoning field (empty content) — the
guard keys on the finish_reason/tool_calls mismatch, so both are covered.
A genuine finish_reason="stop" text turn is unaffected.
Verified live: a scheduled reviewer that died at tool_turns=0 every run
now recovers through 20+ stalls per run and submits real reviews.
Tests: tests/run_agent/test_dropped_tool_call_recovery.py covers the
re-prompt, the empty-content case, the clean-stop control, and the
bounded-loop guard.
After a tool call completes, the conversation loop does to
start the next API call. Between the last (from tool
completion in tool_executor.py) and the next one (at the start of the
next API call), there is a gap. If anything during this gap takes time
— context compression, slow provider prefill, or other post-tool
processing — the combined inactivity window can exceed the gateway's
inactivity_timeout (default 120s). The gateway kills the session and the
user sees 'agent never returns a final response', even though the tool
call itself succeeded in 0.1-0.2s.
Fix: call right before so the gateway
sees a fresh timestamp immediately after tool results are posted,
regardless of how long the follow-up API call takes.
Fixes#69559
Verification follow-up for the #51226 salvage: the host call site guarded
select_context with hasattr(), but the ABC defines a default on every
engine, so the built-in ContextCompressor (and any non-implementing
engine) still paid per-request shallow copies of the conversation
history plus a hook call on every provider request. Identity-check the
bound method against ContextEngine.select_context and return the
request untouched — mirroring the existing base-method short-circuit in
_notify_context_engine_turn_complete — so the default path does zero
work, not just produces an identical result.
Adds two pins: the base no-op is never invoked (patched-to-raise base
stays silent), and ContextCompressor.__dict__ contains neither new verb.
Also registers the contributor email mapping for @chaos-xxl.
Addresses the hermes-sweeper review on #51226:
- _apply_context_engine_selection now passes shallow copies of the read-only
conversation_messages / incoming_message to the hook, so an engine mutating
them in place cannot corrupt persisted transcript state (enforces the
request-only contract, not just documents it). Adds a mutation-regression
test asserting persisted history + incoming message are untouched.
- on_turn_complete docstring: scope the coverage claim to the standard
finalization seam. Some abnormal early-return paths (content-policy block,
provider terminal failure) currently persist+return without finalization and
don't emit the hook; documented as best-effort with a shared-seam follow-up,
rather than over-promising a guaranteed callback for every early exit.
- _apply_context_engine_selection: reject an empty list. all([]) is True, so
a [] returned by a failing/buggy engine previously replaced a valid request
with an empty message list the downstream sanitizers can't restore; now it
falls open to the unmodified request (honors the fail-open contract).
Thanks @johnnykor82 for catching this on #41918's review.
- test: empty list keeps the original request (fail-open regression).
- docs: document select_context()/on_turn_complete() in the public
context-engine plugin guide (were still describing only the old contract).
The on_turn_complete() observation hook is the engine's post-turn signal,
so it should receive the completed turn's canonical token usage when the
host has it, not a hardcoded None. Per @johnnykor82's #41918 contract: the
engine uses prompt/completion + cache_read/write/reasoning buckets to judge
how large/expensive the selected context was before the next select_context().
- conversation_loop.py: stash the most recent provider response's usage_dict
(the same canonical shape fed to update_from_response) on the agent as
_last_turn_usage; reset to None at turn start so turns that never reach a
provider response (early failure / interrupt) forward None, not a stale
prior turn's usage.
- turn_finalizer.py: forward agent._last_turn_usage instead of usage=None.
- context_engine.py: document the usage param contract on the ABC hook.
- tests: cover both ends through the real finalize_turn path — completed turn
forwards the full canonical bucket set intact; no-response turn forwards None.
Co-authored-by: johnnykor82 <johnnykor82@users.noreply.github.com>
Adds the post-turn observation verb as the companion to select_context():
an optional, no-op-default on_turn_complete() called once after the
assistant/tool loop finishes, with the finalized transcript snapshot. Lets
an engine ingest/index/summarize the completed turn to inform the next
select_context(). Wired via _notify_context_engine_turn_complete() from
turn_finalizer.finalize_turn(); fail-open, base no-op short-circuited so
non-implementing engines (incl. the built-in compressor) pay nothing.
This is the request-assembly + observation pair from #41918; with this
commit the PR fully subsumes #41918's two hooks (prepare_request_messages
-> select_context, on_turn_complete) rather than only the selection half.
Co-authored-by: johnnykor82 <johnnykor82@users.noreply.github.com>
Adds an optional, no-op-default select_context() hook to the ContextEngine
ABC, called every turn after the request messages are assembled and before
provider dispatch — independent of should_compress(). Lets an engine select
or replace which context enters the prompt for a single request (retrieval,
topic routing, role/branch switching) without mutating persisted history,
removing the need to abuse should_compress()=True as a per-turn callback.
The host call site (_apply_context_engine_selection) is fail-open: a missing
hook, an exception, or an invalid return value leaves the assembled request
untouched. Additive and non-breaking: the built-in compressor and every
existing engine are unaffected.
Consolidates the per-turn request-assembly surface proposed across #41918,
Related: #36765#41918#24949#47109#50053#23837#25115#29370
agent/tool_executor.py's concurrent tool batch checks agent._interrupt_requested
and aborts the wait early; agent/moa_loop.py's _run_references_parallel had
no equivalent, so a MoA-enabled turn blocked on ThreadPoolExecutor.result()
until every reference model finished or hit its own individual
auxiliary.moa_reference timeout -- there was no way for the user to abort a
live turn mid-fanout.
Thread an optional `agent` parameter through aggregate_moa_context ->
_run_references_parallel (used when MoA references run alongside the main
model) and MoAClient/MoAChatCompletions (used when the MoA preset itself is
the acting model), then poll concurrent.futures.wait() in
_REFERENCE_POLL_INTERVAL_S slices instead of blocking on future.result() per
reference, checking agent._interrupt_requested each cycle.
Deliberately scoped to interrupt/cancel only -- no new or changed timeout
value, so this doesn't overlap open PRs #53784/#53875 (which lower the
per-reference timeout default but don't add interrupt support). `agent` is
optional and defaults to None, so any caller that doesn't pass it keeps
today's uninterruptible blocking behavior unchanged.
Adds test_moa_gemini_aggregator_sanitize_uses_real_model: drives a full MoA
tool-call turn (virtual-provider mode) with a Gemini aggregator and asserts
the strict-API sanitize pass is invoked with the resolved aggregator model
(gemini-3-pro-preview), never the virtual preset name once a slot is
resolved — the exact path that stripped extra_content/thought_signature and
made Gemini aggregators 400 (#65092).
Writing the test surfaced a gap in the salvaged #66212 fix: in virtual-
provider MoA mode (provider=moa, no moa_config threaded through
run_conversation) the conversation-loop branch never fired because it only
consulted moa_config. Extend it to fall back to the facade's
last_aggregator_slot — the same source the handle_max_iterations fix uses —
so both MoA entry modes resolve the real aggregator model.
Also adds the contributors/emails mapping for the #15676 credit base.
When MoA mode is active with a Gemini model as the aggregator,
agent.model holds the virtual preset name (e.g. "closed"), not the
actual aggregator model name. The _sanitize_tool_calls_for_strict_api
call uses agent.model to decide whether to keep extra_content
(thought_signature) on tool_calls — since "closed" doesn't contain
"gemini", the thought_signature is stripped and the Gemini aggregator
rejects the next request with HTTP 400 (INVALID_ARGUMENT):
"Function call is missing a thought_signature in functionCall parts."
Fix: resolve the actual aggregator model name from moa_config
(conversation_loop) or last_aggregator_slot (chat_completion_helpers)
and pass it to _sanitize_tool_calls_for_strict_api so the
_model_consumes_thought_signature check sees the real Gemini model.
Closes#65092
The phase-1 tool-result prune only runs inside compress(), which fires
near 50% of the context window, so it never triggers on large-window
models; old tool outputs then ride in history and are re-sent every turn.
Add prune_tool_results_only(): the same no-LLM prune on a separate, low
proactive_prune_tokens trigger, run as an elif to the compression branch.
Opt-in (default 0), protects the recent tail by message count.
Add the method to the ContextEngine base as a no-op default so pluggable
engines inherit it safely (the post-tool-call path never AttributeErrors on
a non-built-in engine); the built-in compressor supplies the real prune.
Register both keys under the top-level compression config with defaults and
document them.
Interrupted turns can seed a speculative display token count before the provider receives the request. Restore that display-only seed when interruption wins the race, while preserving completed post-compaction state and treating a successful provider response independently of optional usage metadata.
Constraint: #54776 remains reproducible on current main, while review #4702305384 identifies anti-thrashing rollback as stale and usage receipt as an unreliable response-completion signal.
Rejected: Restore anti-thrashing counters from a preflight snapshot | current main derives their verdict from real provider usage after a completed compaction boundary.
Confidence: high
Scope-risk: narrow
Directive: Keep interrupted preflight rollback display-only, and never infer provider completion from the presence of usage metadata.
Tested: ./.venv/bin/python -m pytest -q tests/run_agent/test_413_compression.py (29 passed); turn-finalizer/conversation-loop tests (31 passed); context-compressor targeted tests (12 passed); infinite-compaction targeted tests (3 passed); ruff; git diff --check.
Not-tested: End-to-end interactive interrupt through CLI or gateway transport.
A lock-loser compression pass returns its input unchanged, which the
automatic compression sites misread as 'cannot compress further': the
preflight loop armed the insufficient-progress blocker, the pre-API gate
burned a shared attempt, and a lock-contended 413/overflow retried into
the attempt cap and returned compression_exhausted — which the gateway
answers with a full session auto-reset (#9893/#35809). A temporary
concurrent-compression defer wiped the session.
Consume the landed #69870 lock-skip signal on every automatic path
(preflight in turn_context, pre-API pressure gate, 413 handler, overflow
handler, post-tool compaction): when a pass no-ops AND the type-pinned
lock-skip flag is set, refund the attempt (never count it toward the cap
or the insufficient-progress blocker), and when the turn cannot proceed
(provider already proved the request does not fit) end it with a soft
compression_deferred result — distinct from compression_exhausted — so
the gateway keeps the session intact and the next message retries after
the concurrent compressor finishes.
The new compression_skipped_due_to_lock() reader is type-pinned
(is True or isinstance(str)) per the MagicMock auto-attribute rule, and
compress_context() now also clears the signal at the very top of every
attempt (per-attempt state rule, #58629/#69853) so a stale value can
never make a later breaker/codex no-op look like lock contention.
Salvaged from PR #49874; rebuilt on main's #69870
_compression_skipped_due_to_lock signal instead of the PR's parallel
_compression_deferred_by_lock triple.
aggregate_moa_context's single max_tokens parameter was applied to
both the reference fan-out (_run_references_parallel) and the
aggregator's own synthesis call_llm. #53580 explicitly removed a
hardcoded cap from the aggregator call because it truncated long
aggregator syntheses; #56756 (reference_max_tokens, added to speed up
the advisor fan-out) reintroduced the same shared cap by passing it to
both calls, silently regressing #53580's fix.
Rename the parameter to reference_max_tokens (matching the caller's
own moa_config key) and stop forwarding it to the aggregator's
call_llm invocation, which now always runs uncapped as intended.
* fix(desktop): hide persisted agent-only history scaffolding
Filter verification-stop nudges and context-compaction handoffs at the
stored-history mapper boundary. Preserve a real reply when a compaction
handoff shares its stored message.
* test(desktop): build persisted E2E sessions through the real agent
Drive tui_gateway.entry over its stdio JSON-RPC transport against the mock
provider, wait for real completion events, and persist normal session history
through AIAgent and SessionDB. Migrate resume and hidden-history coverage,
including real compression and live verify-on-stop scaffolding, then remove
the unused direct SessionDB import scripts.
* fix(desktop): use the provisioned Python for real-session E2Es
Run the stdio gateway through uv's synced project environment outside the
Nix dev shell, while retaining the fully provisioned Nix Python when the
shell advertises HERMES_PYTHON_SRC_ROOT.
* fix(nix): expose the provisioned Python environment to uv
Mark the Nix-built Python environment active in the dev shell so the shared
E2E session builder can always run through `uv run --active --no-sync`.
* fix(timeline): persist typed display events
* fix(timeline): strip display-only fields from provider payloads, preserve through rewrites, fix /resume display history
Three review findings from PR #69771:
1. Provider payload leak: display_kind and display_metadata were forwarded
to the provider API as unknown message fields. Strict OpenAI-compatible
backends can reject the next request after a model switch or resumed
typed event. Strip both from the per-request api_msg copy in
conversation_loop alongside the existing api_content pop.
2. Rewrite/import data loss: _insert_message_rows preserved display_kind
but silently dropped display_metadata. After replace_messages,
archive_and_compact, or session import, async-delegation completion
events lost their task counts and fell back to generic display text.
Add display_metadata to the INSERT columns and bind tuple.
3. CLI /resume stale recap: startup --resume A set _resume_display_history
from A's lineage. A subsequent in-session /resume B loaded B only into
conversation_history via get_messages_as_conversation, leaving the stale
A display projection. _display_resumed_history preferentially read the
stale attribute, showing A's recap for B. Switch /resume to
get_resume_conversations and update _resume_display_history alongside
conversation_history.
Tests: 890 Python (5 files), 35 desktop TS — all green.
* feat(tui): render typed display events as ◈ markers in the Ink TUI
The TUI was not handling display_kind at all — model switch markers and
async delegation completions rendered as opaque user messages with the
full [System: ...] text, and hidden compaction handoffs were visible.
Wire display_kind through the full TUI chain:
- _history_to_messages (tui_gateway/server.py) forwards display_kind
and display_metadata to the gateway transcript payload.
- GatewayTranscriptMessage (gatewayTypes.ts) gains both fields.
- Msg.kind (types.ts) gains 'event' value.
- toTranscriptMessages (domain/messages.ts) maps:
- hidden → skip entirely
- model_switch → event "model changed"
- async_delegation_complete → event "N background agents finished"
(or "background agent work finished" without metadata)
- messageGroup (blockLayout.ts) routes event to its own group, with
SELF_SPACED + PAINTS_TRAILING_GAP so it owns its margins.
- messageLine.tsx renders event-kind as a dim ◈ marker with no gutter,
matching the CLI's ◈ event rendering.
- 4 new TUI tests for hidden/model_switch/async_delegation mapping.
TUI typecheck: clean. TUI lint: 0 errors (2 pre-existing warnings).
TUI tests: 9 passed (1 pre-existing failure on main, unrelated).
The dedup-reset calls assumed a full AIAgent; gateway/loop test doubles
built via object.__new__ lack _clear_context_overflow_warn and crashed
in build_turn_context (caught by test_api_content_sidecar on CI slice 3).
getattr-guard all four call sites per the established test-double pitfall
pattern (AGENTS.md #17).
Follow-up fixes for the #62625 salvage:
- Dedup-reset gap (sweeper review): when the block clears while the
context is STILL over threshold, execution enters the compression
branch — the PR's 'else' reset never ran, so the warning stayed
suppressed forever after the first block. _clear_context_overflow_warn()
now fires on every automatic compression path: turn-context preflight,
conversation_loop pre-API gate, and the post-tool loop-compaction gate.
- should_compress_info on current main: main refactored should_compress
into _automatic_compression_blocked()/_locally(); the tuple variant now
derives its reason from the same in-memory state via
_compression_block_reason(), keeping cooldown:<s>/ineffective shapes.
- ContextEngine.should_compress_info ABC default now actually returns
(should_compress(tokens), None) — the PR's default had a docstring but
no return (returned None, would crash tuple-unpacking call sites).
- Below-threshold guard: the turn-context persisted-cooldown branch and
the conversation_loop pre-API cooldown branch no longer warn when the
estimate is under threshold (should_compress_info returns a None
reason; the preflight pre-check is not a threshold guarantee). The
pre-API guard also honors compression.max_attempts instead of a
hardcoded 3, and no longer fabricates a cooldown reason.
- Noise-filter survival (#69550 composition): warning text is now a
template constant (CONTEXT_OVERFLOW_BLOCKED_WARNING_TEMPLATE) marked
FAILURE-CLASS, pinned un-swallowed in VISIBLE_COMPRESSION_MESSAGES and
in new tests that execute the real _TELEGRAM_NOISY_STATUS_RE +
_prepare_gateway_status_message.
- Contributor mapping for stanislav@local -> sl4m3.
- ContextEngine.should_compress_info() default impl so plugin engines
(e.g. _StubEngine) don't raise AttributeError at the call site.
- Centralise warning/reset in AIAgent._warn_context_overflow_blocked /
_clear_context_overflow_warn so turn-context and conversation-loop guards
share identical dedup logic and reset on the real compression boundary.
- Cover conversation_loop.py pre-API (~L1007) and loop-compaction (~L4774)
guards, not just the turn-context preflight.
- _FakeAgent mirrors the two helpers; test suite green (219 passed).
Fixes#62708
Follow-up for the salvaged #35191: the mid-turn pre-API pressure emit in
conversation_loop.py and the idle-resume emit in turn_context.py were not
routed through automatic_compaction_status_message, so an engine with
emit_automatic_compaction_status=False still leaked those lines. Both now
resolve through the hook (phases "pre_api" and "idle") while keeping the
#69550 template constants as the default wording. Suppression also skips the
#69546 structured 'compacted' terminal edge for compress-phase events that
opened no visible phase; failure warnings (_emit_warning) remain never
suppressible, pinned by test.
Invalid-tool exhaustion and truncated-tool early returns skipped finalize_turn, leaving role=tool transcripts that become tool→user on the next turn for strict providers. Call close_interrupted_tool_sequence before persist on those paths (same as interrupt aborts).
Post-sweep audit of every compression status emission (conversation_loop,
turn_context, conversation_compression): the buffered overflow/attempt-cap
retry chatter (🗜️ 'Context too large…', 'Compressed X → Y, retrying…',
'Context reduced to…'), the #69332-reworded auto-lower notice
("Auto-lowered this session's threshold…"), the aux-provider-unavailable
notice, and the concurrent-compression skip all leaked past
_TELEGRAM_NOISY_STATUS_RE on chat platforms. Add anchored alternatives for
each; the ', retrying'/'— compressing' anchors keep manual /compress
feedback ('Compressed: 30 → 12 messages') and failure/abort notices
visible per the deliberate carve-outs.
Also extract every routine compression status string into importable
template constants in agent/conversation_compression.py (single source of
truth shared by all emission sites), so tests can iterate the actual
emitted wording instead of hand-copied literals.
Detect a billing wall once (agent/error_classifier → FailoverReason.billing) and
map it to a recovery link + label in one place, then carry that structured
BillingBlock to every surface instead of re-parsing free-form error text per
surface.
- agent/billing_links.py: provider-agnostic slug/host → (label, billing URL)
table (single source of truth), Nous-aware (is_nous routes to the in-app flow);
unknown providers degrade to a readable label with no invented URL.
- conversation_loop: both billing exit paths return a billing_block through one
helper; the guidance message carries the derived URL for every provider.
- gateway forwards billing_block on message.complete (it was dropped).
- @hermes/shared: BillingBlock type shared by desktop + TUI.
A follow-up sent while the model is still generating previously ended the
turn: Hermes kept only the visible partial text (reasoning was display-only),
cleared the loop, and replayed the message as a fresh next turn. If the
correction referred to something that only appeared in the thinking stream,
the model no longer had that context.
Add `AIAgent.redirect(text)`: a corrective interrupt distinct from a hard
stop. It cancels only the in-flight model request (not tool workers or child
agents), stashes the correction under a lock shared with `interrupt()` so a
concurrent `/stop` always wins, and lets the loop rebuild the same logical
iteration. `_apply_active_turn_redirect()` checkpoints the reasoning that was
actually shown to the user plus any visible partial text as an ordinary
assistant message, then appends the correction as a real user turn — never
replaying incomplete signed/encrypted provider reasoning, and keeping strict
role alternation and prompt-cache stability intact. During tool execution it
degrades to `steer()` so a running tool finishes at a safe boundary.
`_fire_reasoning_delta` now only records reasoning that a display callback
actually consumed, so `show_reasoning: false` never leaks hidden provider
thinking into the persisted transcript.
Follow-up to the salvaged #64010 (Kenmege) and #63870 (dombejar) commits,
making one resolved compression.max_attempts cap govern ALL per-turn
compression attempt sites:
- conversation_loop: resolve max_compression_attempts ONCE at turn start
(it was previously re-resolved inside the API-call loop) and route the
pre-API pressure gate through it — that gate still hardcoded
'compression_attempts < 3' and logged 'attempt=%s/3'.
- conversation_loop: the salvaged post-tool compaction gate now uses the
resolved cap instead of a hardcoded 3.
- turn_context: the preflight compaction loop was 'for _pass in range(3)';
it now sizes itself from the same resolved cap.
- agent_init: harden the max_attempts parser — reject booleans (bool
subclasses int; 'true' would coerce to 1), reject fractional floats
instead of truncating them, keep accepting integral floats and numeric
strings; anything else falls back to 3 (floor 1, ceiling 10 unchanged).
- tests: replace #63870's inspect.getsource source-shape test with
behavioral loop tests (post-tool compaction fires <= cap times per turn,
shares its budget with the pre-API gate, resets between turns); add an
e2e test proving a 4th preflight pass runs at config cap=6 while the
unset default still stops at 3; extend the #64010 config tests with the
bool/float parser semantics.
Salvages #64010 by @Kenmege and #63870 by @dombejar.
The conversation loop hardcodes max_compression_attempts = 3. Sessions
that legitimately need more rounds are stranded: on a restart history
reload, incompressible tool schemas can keep the per-request estimate
above the compressor threshold even though the message floor compresses
correctly, so three rounds cannot clear it and the turn dies with
"Context length exceeded: max compression attempts (3) reached" — the
same failure class as #62605, where the rough estimate similarly leaves
3 retries short.
Make the cap a config key, compression.max_attempts:
- default 3 = identical to today, so an unset key is behavior-neutral;
- parsed and validated in agent_init alongside the other compression.*
keys (>= 1, hard-capped at 10, non-integer values fall back to 3),
attached as agent.max_compression_attempts;
- the loop reads it via getattr(agent, "max_compression_attempts", 3),
so objects without the attribute keep the prior behavior;
- documented in the DEFAULT_CONFIG compression block.
Tests pin the parse/validate/attach seam: default preserved, custom
value honored, floor and ceiling enforced, garbage tolerated, and the
loop-side getattr degradation.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Fix#68178
The git-install auto-updater rewrites source while the desktop backend
is live. Because agent/conversation_loop.py is imported lazily on the
first API call, a process can end up running two different commits
spliced together — one commit's AIAgent against another commit's
conversation_loop. When the interface differs, every turn fails
permanently with an AttributeError, and the loop retries indefinitely,
burning provider API calls (576 failures, 149 wasted API calls observed).
Three-prong fix:
1. Circuit-break AttributeError on agent objects: the outer-loop error
classifier now detects AttributeError targeting agent/run_agent
modules and breaks immediately instead of continuing the retry loop.
2. Code skew detection for desktop/serve backend: run_agent.py now
snapshots the checkout revision at import time and exposes a cheap
per-iteration check that the conversation loop uses to refuse new
work with a clear 'restart required' message before the lazy import
can crash.
3. Informative error message: when code skew is detected, the user
gets a clear explanation of the mismatch (boot revision vs current
revision) and actionable guidance to restart the application.
* fix(desktop): preserve interim assistant text wiped at message.complete
When the agent emits interim text (commentary alongside tool calls, or the
attempted final answer before a verify-on-stop nudge), all UI surfaces
streamed it live but then wiped it at message.complete — keeping only the
final response. The user saw text appear during inference, then disappear.
This is the complete fix across all three layers: agent core, gateway
transport, and all UI surfaces (desktop + Ink TUI).
The verify-on-stop and pre_verify paths flagged the assistant's attempted
final answer as _verification_stop_synthetic, suppressing it from both
state.db and the UI. The user only saw the terse post-verification reply.
Now the assistant response is real content: it's persisted to state.db and
emitted as an interim message via _emit_interim_assistant_message(force_display=True)
before the verification loop runs. Only the synthetic nudge messages keep
the synthetic flags. The turn finalizer drops nudges from live history and
compares content (not just role) to avoid duplicating a published candidate.
Message sequence repair collapses verification candidates in the
consecutive-assistant merge.
Wire agent.interim_assistant_callback both at construction (_agent_cbs())
and per-turn (defense-in-depth), emitting a new message.interim event with
{text, already_streamed}. Gated on display.interim_assistant_messages
(default true). Cleared in the finally block so a stale closure can't
fire on a later turn.
Add message.interim to the GatewayEventName union (apps/shared) and a
typed payload to the TUI's GatewayEvent discriminated union.
The TUI already had the segment-anchoring machinery (flushStreamingSegment +
finalTail) but had no handler for message.interim. Added recordInterimMessage
+ interimBoundaryIndex to seal segments mid-turn, and updated
recordMessageComplete to only dedupe segments after the interim boundary.
Replaced the fragile sealed-set approach with a proper interimBoundaryPending
state flag on ClientSessionState. finalizeInterimAssistantMessage finalizes
the streaming bubble in place (or creates a standalone one), rotates the
stream ID so next deltas create a new bubble, and sets the flag. When the
final text equals an already-sealed interim, they stay as distinct messages.
Extracted mergeFinalAssistantText() as a pure function in chat-messages.ts,
used by both completeAssistantMessage and finalizeInterimAssistantMessage.
Split the bidirectional dedup predicate: reasoning is a restatement only when
the final FULLY covers it. A short final ("Done.") no longer swallows a
longer reasoning block that merely starts with it.
Honor display.interim_assistant_messages (default true) across all layers:
the tui_gateway gates the callback, the desktop wires it to a nanostores
atom via use-hermes-config. Updated hermes_cli/config.py and
cli-config.yaml.example comments to document the Desktop behavior.
_split_segment_tokens now accepts posix=False and _find_ad_hoc_match tries
both posix modes so ad-hoc verification scripts with Windows backslash
paths are matched correctly. (response_previewed forwarding from #53553
is not included — our emit-interim + persist approach makes it unnecessary
since the attempted answer is now surfaced before the verification loop.)
- tsc: clean (desktop + TUI + shared)
- vitest desktop: 73/73 pass (7 interim-sealing + 5 mergeFinalAssistantText + 4 config atom)
- vitest TUI: 83/83 pass (4 new message.interim tests)
- python: 390 tests pass (340 tui_gateway + 33 verification/finalizer + 6 config gating + 3 evidence + 8 continuation budget)
Co-authored-by: Liam Zhang <yingliang-zhang@users.noreply.github.com>
Co-authored-by: Lucas D'Alessandro <lucasfdale@users.noreply.github.com>
Co-authored-by: Eric Manganaro <superposition@users.noreply.github.com>
Co-authored-by: sweetcornna <sweetcornna@users.noreply.github.com>
Co-authored-by: DECK6 <DECK6@users.noreply.github.com>
Co-authored-by: matantsevs <matantsevs@users.noreply.github.com>
Co-authored-by: gitcommit90 <gitcommit90@users.noreply.github.com>
* fix: prefix-match interim streamed content to avoid benign duplicate bubbles
_interim_content_was_streamed used exact equality (streamed == visible_content),
so a final response that was the streamed text plus a trailing delta — or a
partial stream before the verify nudge fired — failed the match and left
_response_was_previewed false. The turn then showed two bubbles (interim +
identical final) instead of settling the interim in place.
Relax to a prefix check (visible_content.startswith(streamed)) in both the
core match and the desktop's settle-in-place gate. The TUI already used
prefix matching via finalTail. The reverse direction (streamed longer than
final) is intentionally not matched — that could suppress a needed resend
in the gateway path where already_streamed=True calls on_segment_break().
* test(desktop): add partial-stream-then-nudge dedup edge case
Third edge case for the interim-sealing dedup: model streams part of its
answer via message.delta, verify nudge fires, interim seals the streamed
prefix, then the final response is the same text plus a trailing delta.
Asserts one bubble (not two) containing the full final text.
Acceptance protocol #2 — covers all three dedup edges:
1. interim == final (existing)
2. interim = strict prefix of final (existing)
3. partial-stream-then-nudge (this commit)
---------
Co-authored-by: Liam Zhang <yingliang-zhang@users.noreply.github.com>
Co-authored-by: Lucas D'Alessandro <lucasfdale@users.noreply.github.com>
Co-authored-by: Eric Manganaro <superposition@users.noreply.github.com>
Co-authored-by: sweetcornna <sweetcornna@users.noreply.github.com>
Co-authored-by: DECK6 <DECK6@users.noreply.github.com>
Co-authored-by: matantsevs <matantsevs@users.noreply.github.com>
Co-authored-by: gitcommit90 <gitcommit90@users.noreply.github.com>
Every API iteration computed `total_chars = sum(len(str(msg)) ...)`, which
str()-serializes the ENTIRE history — including base64 images and large tool
results — just to take its length, then called estimate_request_tokens_rough,
which walked the messages a SECOND time (it re-runs estimate_messages_tokens_rough
internally, already computed one line above).
Now derive both from one image-stripped message estimate:
approx_tokens = estimate_messages_tokens_rough(api_messages) # once
request_pressure_tokens = approx_tokens + tools_tokens # == old value
total_chars = approx_tokens * 4 # log/metric only
request_pressure_tokens is byte-identical to the old
estimate_request_tokens_rough(api_messages, tools=agent.tools or None) (no
system_prompt arg → messages + tools). total_chars only feeds a verbose log and
the pre-api-request hook's request_char_count, so a rough proxy is fine and it no
longer balloons on image turns. On the TTFT critical path for every call.
tests/agent/test_model_metadata.py + test_compressor_image_tokens.py green.
The first LLM call of every gateway turn gets ~0% provider prompt-cache
hit rate (in-turn calls: 97-99%) because the bytes sent for a turn's
user message are not the bytes replayed next turn: memory-prefetch and
pre_llm_call context are injected into the API copy only, the #48677
persist override writes cleaned content to the DB row, and
get_messages_as_conversation sanitize/strips user and assistant content
on load. Any of these diverges the request prefix at that message and
re-prefills everything after it — measured 27.9s for the first call vs
2.4-5.8s cached at a median ~156k-token context.
Persist what you send: a nullable messages.api_content column stores the
exact content string sent to the API when it differs from the clean
stored content, and replay substitutes it verbatim (no sanitize, no
strip). The injection composition lives in one helper
(turn_context.compose_user_api_content); the turn prologue stamps its
output onto the live user message, the api_messages build sends the
stamped bytes, and every outgoing copy pops the field so it never
reaches a provider. The crash-resilience user-turn persist moves after
prefetch/pre_llm_call so the user row is written once with its final
sidecar; _ensure_db_session stays before preflight compression (session
rotation needs the parent row under PRAGMA foreign_keys=ON). The
current-turn index trackers are re-anchored after compaction rebuilds
the message list, in-place preflight compaction backfills the stamp onto
the already-inserted row, and gateway replay forwards the sidecar only
when the replay pipeline did not rewrite the content. Rewrite paths that
would leave stale bytes (historical image strip, merge-summary-into-
tail, consecutive-user repair merge, stale-confirmation redaction) drop
the sidecar; the chat-completions transport and the max-iterations
summary path strip it defensively. codex_app_server and MoA turns are
excluded from stamping because their wire bytes differ from the
composition. A missing or dropped sidecar degrades to today's behavior
(one cache-boundary miss), never to wrong content.
1. Remove PR's redundant strip_think_blocks coercion (Teknium's fix
296494db0 already handles this at the same chokepoint with superior
logic that drops thinking/reasoning blocks).
2. Restore 'if not content: return ' guard at top of strip_think_blocks
that was lost during cherry-pick auto-merge. Without it, None content
hits str(None) → 'None' string instead of returning empty.
3. Fix error classifier design flaw: remove 'conversation_loop' and
'run_agent' from _local_processing_modules — these are the container
modules for the try/except, so every exception passes through them,
making _hit_local always True and misclassifying transient API/network
errors as non-retryable local bugs.
4. Move module sets to module-level frozenset constants (_LOCAL_PROCESSING_MODULES,
_API_CALL_MODULES) instead of rebuilding on every exception.
5. Replace traceback.extract_tb() with raw tb walk — avoids disk I/O for
source lines that are never used.
6. Remove unused 'import traceback'.
7. Fix docstring corruption: 3 lines where think tags were replaced
with Chinese characters during the PR's editing.
Degrading models (observed with gpt-5.6 past ~350K input) emit tool-call
batches like 6 valid named calls + 1 blank-name call. Previously the
whole turn was voided — every valid call got 'Skipped: another tool call
in this turn used an invalid name' — and three such batches tripped the
3-strike stop, killing sessions that were still making progress.
Now a mixed batch error-results ONLY the invalid call(s) (terse
anti-priming error for blank names per #47967, catalog dump for typos)
and dispatches the valid subset for execution. The assistant message
keeps every emitted call so provider-side tool_call/result pairing stays
intact. The 3-strike counter only advances when a turn contains NO valid
call, so a fully-degenerate model still stops while a mostly-coherent
one keeps working. Broken JSON args on a never-executing invalid call no
longer trigger the whole-turn JSON retry loop.
Field evidence: July 2026 debug bundle showed gpt-5.6-sol emitting
6-call batches with one blank-name rider at 559K/384K-token context in
two separate sessions; 13 valid tool calls were discarded before the
session stopped as partial.
Three fixes for the Bedrock Claude path:
1. Streaming fallback: When AnthropicBedrock SDK raises 'Unexpected event
order' (SDK misparses Bedrock error events as message_start), auto-switch
to native Converse API for the rest of the session instead of failing
after 3 retries.
2. Image base64 decode (#33317): data URL payloads were passed as base64
strings to source.bytes, but boto3 re-encodes at the wire layer. Now
decoded to raw bytes before passing to Converse API.
3. Bearer token routing (#28156): Users with AWS_BEARER_TOKEN_BEDROCK are
now routed through Converse API regardless of model, since the
AnthropicBedrock SDK only supports SigV4 signing.
3 new tests. 121 bedrock_adapter tests passing.
Two consecutive incomplete assistant interims with identical visible
content (content + reasoning) are collapsed even when opaque provider
state (encrypted reasoning item ids, message item phases) drifts per
continuation — previously that drift defeated dedup and caused message
storms (#52711). The latest opaque payload is written onto the existing
message in place, so continuation replay still uses fresh provider
state.
Salvage note: the original PR also made the 3-retry continuation cap
cumulative per turn (no reset on progress). That half is intentionally
dropped — a legitimate long turn alternating incomplete/progress would
hard-fail at 3 cumulative, changing semantics beyond the reported bug.
Map Codex Responses status=incomplete with incomplete_details.reason=content_filter to finish_reason=content_filter so the existing refusal/fallback path runs instead of burning incomplete continuation attempts.
Community reports of GPT 'infinitely thinking' are usually a slow or
overloaded provider plus silent retry machinery: the CLI/TUI/Desktop
spinner shows a generic 'cogitating...' verb for the whole wait and the
gateway heartbeat says only 'Working — N min'.
Add AIAgent._emit_wait_notice(): rewrites the live spinner/status line
(thinking_callback → CLI prompt_toolkit widget, thinking.delta → TUI +
Desktop) and updates the activity tracker (included in the gateway's
'⏳ Working — N min' heartbeat). Wired at the four wait points:
- non-streaming wait loop: after 30s with no response, the line becomes
'⏳ waiting on <model> — Ns with no response yet (provider may be slow
or overloaded; auto-reconnect at Ns)'
- streaming wait loop: same explanation after 30s with no chunks,
including the long-thinking case
- TTFB / stale-stream kills: '⚠ no response from provider in Ns —
reconnecting...'
- Codex continuation retries: '↻ model returned reasoning with no final
answer — asking it to continue (n/3)' instead of silence
Notices are fail-open (display errors never break the wait loop) and
gateway sessions without a display callback still get the improved
activity description.
Follow-up to the salvaged #63690: when the interim assistant message is
too empty to append (no content and no reasoning of any kind), the last
message in history is still the prior user/tool turn — appending the
user-role nudge there would create a user→user or tool→user sequence
that strict providers reject. Only append the nudge when the last
message is an assistant turn. Also add IpastorSan to AUTHOR_MAP.
grok-4.x on the xAI /v1/responses surface sometimes ends a turn with only
reasoning items — no message output item, no tool calls — and those
reasoning items carry no encrypted_content. Two compounding problems:
1. The model occasionally emits its final answer INSIDE the reasoning
channel, delimited by grok's internal "<response>" tag. The answer
exists but is classified reasoning-only → finish_reason=incomplete.
2. An interim assistant message holding only plain-text reasoning replays
as nothing in _chat_messages_to_responses_input, so every continuation
request is byte-identical to the one that just failed. The model
deterministically repeats the reasoning-only response until the retry
budget is exhausted and the turn dies with "Codex response remained
incomplete after 3 continuation attempts".
Fixes:
- _normalize_codex_response (xai_responses only): salvage the
<response>-delimited tail from the reasoning text and promote it to
assistant content; the untagged prefix stays as thinking text.
- Codex-incomplete continuation path: when the interim message has
nothing the input converter will replay (no content, no encrypted
reasoning items, no message items), append a user-role nudge so the
retry actually differs and explicitly asks for the final answer /
pending tool call. Mirrors the existing _get_continuation_prompt
pattern used for length truncation.
Observed live with grok-4.20 on xai-oauth (2026-07-13); sibling of the
grok-composer web_search incomplete-loop fix in transports/codex.py.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The upstream migration to inference.nousresearch.com broke routing for
inference-api.nousresearch.com. Restore the correct base_url and drop
the stale alias match in _is_nous_inference_route().
Fixes#60715
Add a bounded turn-end stop guard for kanban workers. When a worker
tries to exit with finish_reason=stop without having called
kanban_complete or kanban_block, inject up to two synthetic nudges
so the conversation loop continues instead of exiting cleanly (which
the dispatcher records as protocol_violation).
Mirrors the existing verify-on-stop pattern: same ephemeral scaffolding
flag (_kanban_stop_synthetic), same role-alternation contract, same
_pending_verification_response fallback for budget exhaustion.
Disabled by default (gated on HERMES_KANBAN_TASK env var set by the
dispatcher); kill switch via HERMES_KANBAN_STOP_NUDGE=0.
Salvaged from #62262 by @mdc2122. The original branch was 272 commits
behind main with ~538 files of stale-base reversions; this salvage
applies only the 4 substantive files (agent/kanban_stop.py,
conversation_loop.py insertion, run_agent.py _EPHEMERAL_SCAFFOLDING_FLAGS,
tests/agent/test_kanban_stop.py).
Salvage of #63888. The original fix clears stale _last_content_with_tools
on substantive tool-only turns but doesn't clear _mute_post_response, which
a prior housekeeping turn may have set. This suppresses tool progress
output via _vprint until the no-tool-call branch resets it at line ~4834
— after all tools have finished executing.
Fix: also reset _mute_post_response = False when clearing stale fallback.
Added test: verify pure housekeeping turns (content + only housekeeping
tools) still set the fallback correctly — the original use case the
fallback was designed for.
Co-authored-by: liuhao1024 <sunsky.lau@gmail.com>
A cached _last_content_with_tools response from a housekeeping-only turn
could survive a later substantive tool-only turn. When the model returned
an empty response, Hermes incorrectly finalized the older housekeeping
narration instead of invoking the post-tool empty-response nudge.
Production impact: scheduled cron jobs could return early without completing
their actual work (e.g., daily report job returning a housekeeping message
instead of producing the report artifact).
Root cause: The fallback state was only updated when a turn had both
content AND tool_calls. A turn with tool_calls but empty visible content
would skip state updates entirely, leaving stale fallback state intact.
Fix: Classify tools in every tool-call turn (regardless of visible content).
When any tool is substantive (non-housekeeping), clear the older fallback state
before processing later empty responses. This prevents two-turn-old housekeeping
narration from being treated as if it belonged to the immediately preceding
substantive tool turn.
Regression test added: tests/run_agent/test_conversation_fallback_state.py
Fixes#63860
Cache-decorated turns (apply_anthropic_cache_control converts string
content to [{type: text, ..., cache_control}] lists — applied BEFORE the
MoA facade since the #57675 cache-cold fix) and multimodal turns
(text + image_url parts) flattened to empty strings in
_reference_messages, which only read str content. On turn 1 of a
provider:moa session with a Claude aggregator the references received a
single EMPTY user message: Anthropic-side providers 400'd ('messages: at
least one message is required') while tolerant models answered 'no user
request is present' (live incident Jul 14 2026, preset 'closed').
Fixes, in totality:
- _reference_messages: extract visible text via
agent/message_content.flatten_message_text for user/assistant/tool
turns (skips image parts, so no base64 leaks into the advisory view);
decorated and undecorated transcripts now produce a byte-identical
advisory view (advisor cache prefix stays stable).
- image-only user turns get a placeholder instead of an empty message
(Anthropic rejects empty text blocks) or a silently dropped turn
(would break user/assistant alternation).
- degenerate-case fallback flattens structured content too.
- _attach_reference_guidance: a decorated/multimodal trailing user turn
now receives the guidance as a NEW text part appended AFTER the
cache_control-marked part (cached prefix byte-stable) instead of
falling through to a second consecutive user message (strict providers
reject user/user).
- conversation_loop MoA injection: multimodal user turns get the MoA
context appended as a trailing text part instead of being dropped;
user_prompt for the one-shot path flattens content lists instead of
str()-ing them (which leaked base64 payloads into the prompt).
Live-verified on the 'closed' preset (real OpenRouter wire, 2 user
turns, tool loop): all 4 reference calls carry the full document +
rendered tool state, end on user, zero tool-role/tool_calls; advisor
cache_write 7968 then cache_read 5909+; aggregator cache_read
14880-15237 on iterations 2+.
Co-authored-by: bo.fu <bo.fu@meituan.com>
Salvage of #63862. is_output_cap_error() returns False for vLLM/LM Studio
error messages that contain 'prompt contains ... input tokens' (treated as
input-overflow signal). But parse_available_output_tokens_from_error() CAN
extract a valid available_tokens from those same messages. The
compression-disabled guard only checked is_output_cap_error(), so vLLM/LM
Studio users with compression off still got a terminal failure instead of
the max-tokens retry.
Fix: also exempt when parse_available_output_tokens_from_error() returns a
value — that function determines whether the retry path can actually handle
the error, so it's the right predicate for the exemption.
Added test: verify vLLM-format error with compression_disabled=False still
triggers the max-tokens retry path.
Co-authored-by: dmabry <dmabry@users.noreply.github.com>
The branch computed safe_out from estimate_messages_tokens_rough(messages),
but the provider rejected the larger api_messages request (system prompt,
injected context, tool schemas). When API-only content is large, safe_out
could far exceed the provider's available_tokens.
Compute safe_out from estimate_request_tokens_rough(api_messages, tools=...)
and keep provider available_out as an upper bound. Do not alter context_length
or trigger compression for output-cap errors.
Add production-path run_conversation tests that assert the retry API call's
max_tokens, including a case where a large system prompt makes messages-only
estimation undercount the real request.
Fixes#55546
The retry loop computed safe_out from the error's available_tokens,
which reflected the *previous* request. Between retries the agent
appends tool results and error text, so the real input token count
grows. Deriving safe_out from the stale budget meant every retry
still exceeded the context ceiling by 1+ tokens, burning through the
3-attempt limit.
Compute safe_out from estimate_messages_tokens_rough(messages) so
the cap tracks the growing input on each retry attempt.
Cron jobs in the gateway process wedged before HTTP on later non-streaming
API calls because interruptible_api_call spawned a daemon worker inside
nested cron thread pools. Route cron platform turns through direct_api_call
on the conversation thread instead.
Treat intent-ack continuation text as non-final so last-turn exhaustion requests a real summary instead of surfacing a premature promise. Keep iteration-limit fallback text free of the abnormal-fragment explainer.
Name the continuation fallback for its actual verification-only provenance so unrelated continuation paths cannot accidentally inherit its cron-delivery semantics.
Track held-back verification responses explicitly so budget exhaustion returns the composed report without a second model call. Keep unrelated error and recovery exit reasons intact, preserve Kanban timeout accounting, and cover the real run_conversation paths.
Clear stale final_response before verify-on-stop/pre_verify loop
continues, and normalize iteration-limit exit reasons in turn_finalizer
when a composed answer survives with unknown/budget_exhausted.
Fixes#61631
Follow-up on the salvage of #59523. Two low-risk cleanups surfaced by review:
- Extract _ZAI_CODING_OVERLOAD_SHORT_ATTEMPTS as a module constant so
adaptive_rate_limit_backoff() and zai_coding_overload_retry_ceiling()
share one source of truth. Previously both hardcoded short_attempts=3
independently; tuning one without the other would silently desync the
retry ceiling from the backoff schedule.
- Replace the tautological formula-mirroring assert in
test_zai_overload_retry_ceiling_exceeds_short_attempts with a behavior
invariant (ceiling leaves headroom for every long-backoff entry), per the
repo's contracts-over-snapshots testing rule.
Z.AI Coding Plan GLM-5.2 reports server overload as HTTP 429 code 1305
("temporarily overloaded"). classify_api_error routes that to
FailoverReason.overloaded (so a valid credential pool isn't burned), but
the adaptive Z.AI backoff was gated on is_rate_limited — which excludes
overloaded — so it never ran (policy=default) and the request failed after
a few quick short retries.
Two compounding causes, both fixed here:
1. Detect the Z.AI overload 429 directly and let its adaptive backoff run
on the overloaded path, not only the rate_limit path.
2. Raise the retry ceiling for this narrow case via
zai_coding_overload_retry_ceiling(). The long-backoff tier
(30/60/90/120s) starts after short_attempts (3) retries, but the default
api_max_retries is also 3, so the loop always gave up before the long
tier could run — leaving the whole long-backoff schedule as dead code.
Scope is limited to the existing narrow is_zai_coding_overload_error match,
so other providers' 429/503/529 handling is unchanged.
The google-antigravity and google-gemini-cli OAuth providers were removed
in #50492. They were the only producers of a cloudcode-pa:// base_url, so
the account-level-quota early-returns in _pool_may_recover_from_rate_limit
and _credential_pool_may_recover_rate_limit are now unreachable.
- Drop the dead cloudcode-pa:// checks and the now-unused provider/base_url
params on _pool_may_recover_from_rate_limit (only caller updated).
- Prune the obsolete CloudCode-specific regression tests; keep the live
single/multi-entry pool-rotation invariants (#11314).
ruff check --fix --select F541 . on current main. Pure prefix removals;
adjacent-string concatenations keep the f only on interpolating fragments.
No string content or live placeholder altered.
Inspired by Claude Code v2.1.199 (July 2, 2026): SSL certificate errors
(TLS-inspecting proxies, missing CA bundles, expired certs) no longer
burn retries before showing actionable guidance — they fail immediately
with the fix hint.
- agent/error_classifier.py: new FailoverReason.ssl_cert_verification +
_SSL_CERT_VERIFY_PATTERNS, checked BEFORE the transient-SSL patterns
(cert-verify messages also contain '[SSL:' and previously retried
forever as timeout). Non-retryable, no compression, no fallback churn.
- agent/conversation_loop.py: dedicated status line + per-cause fix
hints (corporate proxy CA bundle, certifi refresh, self-signed local
endpoints) on the non-retryable abort path.
- 7 new tests incl. regression guards (transient alerts still retry,
large-session cert failure doesn't trigger compression).
* fix(cli): set correct x-initiator header per Copilot turn
copilot_default_headers() always hardcoded x-initiator: agent, but
GitHub Copilot billing requires "user" for user-initiated prompts and
"agent" for tool/follow-up calls. This caused premium requests to never
be consumed correctly, risking billing issues or account bans.
Adds is_agent_turn param to copilot_default_headers() and injects
extra_headers={"x-initiator": "user"} on the first API call of each
user turn when targeting Copilot URLs. The flag flips to False after
injection so subsequent calls (tool use, streaming fallback) default
back to "agent".
Fixes#3040
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* chore(release): add AUTHOR_MAP entry for @tjp2021 (PR #4097 salvage)
---------
Co-authored-by: Tim <tim@iteachyouai.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Self-review (hermes-pr-review Phase 2) flagged the mid-turn pre-API
compaction for reuse/duplication (W1/W2); fixing that surfaced a
regression against a deliberate existing feature, now also fixed.
The block now mirrors the turn-prologue preflight's guard chain exactly
(agent/turn_context.py) instead of a hand-rolled pressure limit:
1. should_defer_preflight_to_real_usage(rough) — defer when the rough
estimate is known-noisy vs a recent real provider prompt that fit
under threshold (schema overhead / post-compaction over-count, #36718).
2. get_active_compression_failure_cooldown() — skip during a same-session
compression-failure cooldown.
3. should_compress(rough) — reuses the canonical threshold_tokens (output
room already reserved by _compute_threshold_tokens) plus its summary-LLM
cooldown + anti-thrash guards (#11529).
Dropped the seven inline _reserve/_output_pressure locals (W2: they
re-derived _compute_threshold_tokens and omitted its 85% degenerate-window
fallback). compression_attempts stays as the hard per-turn backstop.
Without guard (1) the block fired a compaction the preflight deliberately
defers, breaking test_413_compression::test_preflight_defers_when_recent_
real_usage_fit (ValueError from the mocked _compress_context). Verified:
test_413_compression 26/26, codex 82/82, and anti-thrash engaged
(_ineffective_compression_count=2) still suppresses the block (0 calls).
The salvaged max-output/pressure fix set conversation_history=None after
the new pre-API compaction. That is only correct for legacy session-
rotation. Under the default in-place compaction (compression.in_place:
True), archive_and_compact inserts the compacted rows into the session DB
directly without stamping them with the intrinsic persisted-marker, so a
subsequent flush with conversation_history=None re-appends them — doubling
the active context and retriggering compression (the early-persist
duplicate-row trap).
Use conversation_history_after_compression(agent, messages), matching the
two existing compaction sites (post-response should_compress and the
turn-prologue preflight), which returns None for rotation and
list(messages) for in-place so the compacted dicts are skipped by identity.
Adds a regression test with a real SessionDB + real archive_and_compact
that asserts the compacted summary row is persisted exactly once (fails
with 2 copies on the None variant).
A single-model Hermes agent never sends temperature; the provider default
applies. MoA hardcoded reference_temperature=0.6 / aggregator_temperature=0.4,
and the coercion float(preset.get(key, 0.6) or 0.6) made unset IMPOSSIBLE to
express: absent, null, empty, and even an explicit 0 all collapsed to the
baked-in default. Every MoA advisor and aggregator therefore ran at 0.6/0.4
while the same model running solo used the provider default — silently
skewing solo-vs-MoA comparisons and overriding provider-tuned defaults.
- moa_config normalization: temperatures coerce to None when absent/blank/
invalid (new _coerce_float_or_none); explicit values incl. 0 honored.
- moa_loop: _preset_temperature() resolves preset values; None flows to
call_llm, which already omits the parameter when None (same contract as
max_tokens). Aggregator still inherits the acting agent's own configured
temperature when the preset doesn't pin one.
- conversation_loop (context-mode MoA): same resolution, no more hardcoded
0.6/0.4 at the call site.
- DEFAULT_CONFIG preset + web_server payload models + docs updated: unset
is the default, pinning stays available.
MoA per-turn latency is dominated by advisor GENERATION: turn wall time
correlates ~0.88 with output tokens and ~-0.03 with input tokens (measured over
52 turns). Each turn waits for the slowest advisor to finish writing, and
advisors were uncapped — writing multi-thousand-token essays the aggregator
only needs the gist of.
Add an opt-in per-preset reference_max_tokens knob (mirrors reference_temperature)
that caps ADVISOR output only; the acting aggregator is never capped. Default
None = uncapped, so existing presets are byte-for-byte unchanged (no regression).
Wired through both MoA execution paths (MoAChatCompletions.create and
aggregate_moa_context).
E2E: same task, closed preset uncapped vs reference_max_tokens=600 -> 59s to 33s
(~44% faster), final answer identical/correct.
- hermes_cli/moa_config.py: _coerce_int_or_none helper + reference_max_tokens
in _normalize_preset/_default_preset/flattened view
- agent/moa_loop.py: read preset.reference_max_tokens, pass to reference fan-out
- agent/conversation_loop.py: pass reference_max_tokens on the per-turn path
- tests + docs
On the MoA path agent.model/provider are the virtual preset name (e.g.
"closed") and "moa", which have no pricing entry. estimate_usage_cost()
returned None for the aggregator turn, so the `if amount_usd is not None`
guard skipped it and the session's estimated_cost_usd reflected only the
advisor fan-out — a ~50% undercount when the aggregator does the full acting
loop (verified: $0.91 advisor-only vs $1.96 true, aggregator = 54%).
MoAChatCompletions.create() now stashes the resolved aggregator slot as
last_aggregator_slot (exposed via MoAClient); conversation_loop reads it to
price the aggregator turn at its real model/provider. cost_source flips from
'none' to 'provider_models_api'.
Adds Vertex AI as a first-class provider for Gemini models via Vertex's
OpenAI-compatible endpoint. Vertex authenticates with short-lived OAuth2
access tokens (service-account JSON or ADC), not a static API key — the
missing piece behind the recurring requests (#13484, #12639, #56259).
- agent/vertex_adapter.py: OAuth2 token minting + refresh-on-expiry
(5-min margin), ADC->service-account fallback, global vs regional
endpoint URLs. Config precedence: env var > config.yaml > default.
- plugins/model-providers/vertex/: provider profile (auth_type=vertex),
reuses Gemini's extra_body.google.thinking_config translation.
- runtime_provider: vertex short-circuit BEFORE the credential pool so a
credentials-file path is never mistaken for a static API key; mints a
fresh token + computes base_url per resolve.
- run_agent + conversation_loop: _try_refresh_vertex_client_credentials()
re-mints the token and rebuilds the client on a mid-session 401, so a
long-lived gateway agent survives token expiry (~1h).
- auxiliary_client: vertex auth_type branch for side-LLM tasks.
- config.yaml: vertex.project_id / vertex.region (non-secret, bridged to
env); credential path stays in .env (VERTEX_CREDENTIALS_PATH).
- setup wizard + model picker: dedicated _model_flow_vertex; curated
google/gemini-* model list; --provider choices.
- pricing/metadata: Vertex prices off the gemini docs snapshot; endpoint
host auto-maps to the vertex provider (no probe spam).
- lazy_deps + pyproject [vertex] extra: google-auth, opt-in only.
- docs: guides/google-vertex.md + providers page; tests for adapter +
runtime resolution.
Salvages and modernizes #8427 by @slawt onto current main: rewired from
the legacy PROVIDER_REGISTRY path to the provider-profile architecture,
moved non-secret config out of .env into config.yaml, and added the
per-turn 401 token-refresh the original lacked.
MoA full-turn traces (moa.save_traces) recorded the aggregator's acting
output only on the non-streaming path, where it's captured inline at
call time. On the streaming path — which every hermes chat --query run
and every live gateway/CLI turn takes — the aggregator's raw token
stream is handed to the live consumer, so the trace left output=null and
only pointed at the session-db assistant row. An offline audit of a
benchmark run (HermesBench drives --query) then couldn't see what the
aggregator produced without hand-joining to state.db.
Capture the resolved streamed acting text at trace-flush time (the agent
already holds it in _current_streamed_assistant_text) and fold it into
the trace, so the record is self-contained in both modes. New
output_location value inline_from_stream marks a streamed turn whose text
was captured this way; a genuinely empty acting turn (pure tool call)
still points at the session db, matching state.db exactly.
Touches only the trace side-channel — no change to the acting path,
message history, role alternation, or prompt cache.
- agent/moa_loop.py: consume_and_save_trace(..., aggregator_output_fallback)
on both the facade and the MoAClient wrapper; prefer inline capture,
fall back to the resolved streamed text.
- agent/moa_trace.py: embed the fallback; add inline_from_stream location.
- agent/conversation_loop.py: pass _current_streamed_assistant_text at flush.
- tests: 5 cases across streaming / non-streaming / empty-fallback / no-double-write.
teknium1
Drexuxux
konsisumer
Jash Lee
webtecnica
xue xinglong
srojk34
robbyczgw-cla
AlexFucuson9
Kolektori
izumi0uu
helix4u
Alex Fournier
ethernet
Stanislav
Stephen Schoettler
Brett Bonner
Fangliquan
brooklyn!
Jakub Wolniewicz
Dom Bejar
Kennedy Umege
cucurigoo
x7peeps
Soju06
kshitijk4poor
nanami7777777
Eva
JiaDe-Wu
liuhao1024
Sk
Ignacio Pastor
Michael Huang
mdc2122
dmabry
HexLab98
Umi4Life
xxxigm
eliteworkstation94-ai
Steve Lawton
Dutch Dim