A tool-using Transparent Stream turn renders multiple assistant segments
(activity segment(s) then the final answer). Turn metadata — including
_usedModel — is stamped only on the final assistant message, but the
transparent footer read it via turn.querySelector('.assistant-segment'),
which returns the FIRST segment and thus no metadata. Combined with the
generic footer yielding model ownership to transparent mode, the model
label disappeared entirely on exactly the tool turns where per-turn
attribution matters most.
- Add _transparentTurnMetaMessage(turn): scan .assistant-segment
[data-msg-idx] last→first for the final metadata-bearing assistant
message, falling back to the last assistant segment.
- Use it for the transparent footer's duration/model/TTFT/usage instead
of querySelector's first match.
- Non-blocking UX (per gate): preserve the full uncompacted model id in a
hover title on both the transparent .lf-model span and the settled
.msg-used-model-inline chip.
- Tests: behavioral multi-segment harness asserting exactly one .lf-model
(and the correct final-segment metadata + hover title), plus wiring
checks for the segment-scan helper.
Addresses the three gate-review blockers on #6113:
- Fallback attribution: read the served model from agent.model AFTER the
run (the agent mutates it when a fallback fires) instead of the pre-run
resolved_model, for both the persisted _usedModel and SSE used_model.
- Gateway double-label: suppress the additive chip whenever
_gatewayRouting is present, and pass msg._usedModel to
_formatGatewayModelLabel as its model fallback when routing omits
used_model.
- Transparent double-label: skip the generic .msg-used-model-inline chip
when the transparent turn footer owns the model label (.lf-model).
Also per Greptile: settled footer now orders the model chip after the
duration (elapsed - model - ...) to match the transparent footer, and the
test file gains its trailing newline.
Record resolved_model on assistant messages as _usedModel, expose used_model
in the usage SSE payload, and render model chips in transparent and settled
turn footers while suppressing duplicates when gateway routing already shows
used_model.
Per Greptile's review on #5717: the previous
`if result.get('messages') is not None` guard created an asymmetric
condition where the silent-failure classifier could miss the injected
fallback when the agent result had `messages: None` (which never
actually happens in practice since `finalize_turn` always returns a
list, but the asymmetry read as code smell). Drop the guard; the
writeback is now unconditional on `isinstance(result, dict)`.
Per the maintainer's review on #5717: there may be a
`finalize_turn` invariant on the agent branch this WebUI ships
against that already appends the fallback as an assistant row, in
which case the live bug is gone and this PR becomes a regression
guard. Reframe the test docstrings as dual-purpose pins of the user-
visible closure-text contract so they read correctly under both
interpretations.
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
The WebUI already surfaces the model-generated summary when the agent's
tool-iteration budget is exhausted. The remaining gap: when the agent
hits the limit but its summary call yields no usable assistant content
(common with reasoning-only responses), the user sees a bare
`tool_limit_reached` error instead of any closure text.
`hermes-agent.handle_max_iterations()` always returns a non-empty
`final_response` (either the summary or a graceful fallback), and
`AIAgent` exposes that string on the result dict. The WebUI, by
contrast, only reads `result['messages']`, so an empty summary
left no final assistant answer and routed to the error path.
Inject `result['final_response']` as a final assistant turn when
`max_iterations_reached` fires and no usable answer exists in
`messages`. The existing `_mark_latest_assistant_tool_limit_status`
flow then attaches the status card and the user sees the closure
text instead of the bare error. Hermes-agent parity preserved (it's
literally the same string); the bare-error path stays for the case
where the agent died entirely (no fallback either).
Tests cover: fallback-injection surfaces closure text + status card,
no double-injection when a real summary already exists, the existing
'died entirely' apperror path is unchanged, and three unit cases for
the helper's skip conditions.
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
Address round-2 gate cert (issuecomment-4896321817) on sha:dcec232d:
1. BRICK (Codex strace-verified): v2's fcntl-flock holder check on
.git/index.lock was wrong. Codex straced git 2.43.0 and proved
that git uses O_CREAT|O_EXCL + rename(2), NOT advisory locking,
so an fcntl-flock probe returns False for a lock file a live git
process is holding open. The 'fail-closed' claim was wrong:
auto-delete could still race and corrupt the index.
2. CORE (Codex): if .git/index.lock vanished between _is_lock_held
and os.remove, the FileNotFoundError surfaced as a 'remove-failed'
diagnostic instead of 'already gone' and prevented the apply retry.
Fix: stop deleting lock files from the server entirely. The only thing
that removes a lock is the user, on the host, via the manual command
surfaced in the response. Once the lock is gone, the user re-clicks
Update Now and the normal non-destructive apply path runs.
Implementation:
api/updates.py:
- Removed _is_lock_held (fcntl-flock probe) -- was provably unsafe.
- Removed _try_remove_lock (os.remove path) -- was provably unsafe.
- Removed _GIT_LOCK_FILES_REMOVABLE (no longer enumerable by name).
- Added _inventory_locks(path) -> dict: pure inventory of well-known
+ other lock files under .git/. Never touches anything.
- Rewrote apply_clear_lock to inventory-only + manual-instruction.
When .git/index.lock is absent, runs the normal non-destructive
apply path; when present, returns ok=False with the exact 'rm -f'
command + an explanation of why the server cannot do this safely
(O_CREAT|O_EXCL cannot be detected with advisory probes).
CORE-1 race evaporates: there is no os.remove to race against.
api/routes.py:
- Updated the /api/updates/clear_lock comment to reflect v2.2.
static/ui.js:
- applyClearUpdateLock: when res.lock_held, call new
_renderLockManualInstruction(target, res) which surfaces the exact
'rm -f <path>' command in a copyable code block plus an
'I've removed the lock -- retry update' button that POSTs the
same endpoint again; the second call takes the success branch
and re-runs the normal apply path. Also lists any other lock
files present so the operator can investigate.
tests/test_updates.py:
- Removed all 4 tests for the deleted v2 helpers (_is_lock_held
Returns*, _try_remove_lock*).
- Added test_v2_probe_helpers_removed: regression guard that
fails loud if either helper is ever reintroduced.
- Rewrote test_apply_clear_lock_removes_unheld_lock_and_runs_normal_update
and test_apply_clear_lock_refuses_when_lock_held for the v2.2
contract. The 'refuses' test monkeypatches os.remove to record
any deletion attempt and asserts the call list is empty -- a
strict runtime guard against future regressions to auto-delete.
- Added 7 new tests: test_inventory_locks_* (4), test_apply_clear
lock_with_no_lock_runs_normal_update, test_apply_clear_lock_with
lock_present_returns_manual_instruction, test_apply_clear_lock
listing_includes_other_locks, test_apply_clear_lock_rejects
unknown_target, test_apply_clear_lock_rejects_not_git_repo.
Verification:
./scripts/test.sh tests/test_updates.py -> 73 passed
./scripts/test.sh tests/test_update_banner_fixes.py -> 102 passed
./scripts/test.sh tests/test_api_timeout.py -> 8 passed
python3 scripts/ruff_lint.py --diff upstream/master -> 0 new violations
Full ./scripts/test.sh -> 12137 passed, 18 failed (all 18 pre-existing
flakes in unrelated subsystems; verified not v2.2 regressions)
Address Greptile P1 review (discussion r3530941293) on commit e8f7e28a:
the v2 frontend recovery path was dead at the DOM level. static/ui.js
references $('btnClearUpdateLock') but no matching element existed in
static/index.html, so users hitting a lock conflict never saw a clickable
affordance -- the error panel was the only visible state.
Add the missing button next to btnForceUpdate in the update-banner action
row:
<button class="update-btn" id="btnClearUpdateLock"
style="display:none"
onclick="applyClearUpdateLock(this)">
Clear lock and retry update
</button>
Style mirrors the neutral update-btn (NOT the red --error used by the
destructive Force update) so the recovery affordance doesn't read as a
warning. Hidden by default; revealed by the existing _showUpdateError()
helper when res.lock_conflict is set.
Regression tests in TestClearLockButton (3 tests):
- test_clear_lock_button_exists
- test_clear_lock_button_hidden_by_default
- test_clear_lock_button_calls_applyClearUpdateLock_handler
These lock in the contract that the button element must be present and
correctly bound; future refactors that rename the id or strip the button
will fail loud. Mirrors the existing TestIndexHtmlBanner conventions
for btnForceUpdate.
Verification:
- ./scripts/test.sh tests/test_updates.py -k lock -> 35 passed
- ./scripts/test.sh tests/test_update_banner_fixes.py -> 102 passed
- ./scripts/test.sh tests/test_api_timeout.py -> 8 passed
- ruff_lint --diff upstream/master -> 0 new violations
Address the RED gate cert on PR #5688 (2 BRICK + 1 CORE) and the
incoming Greptile review (1 P1 + 3 P2):
- BRICK-1 (race-safety): drop the mtime heuristic and any age-based
lock removal. Add /api/updates/clear_lock with a fail-closed
fcntl.flock holder probe (refuses if any process still holds the
file, on POSIX). On non-POSIX, fails closed. Touches only
.git/index.lock (the only well-known short-lived lock git creates).
- BRICK-2 (destructive path coupling): remove lock_conflict from the
force-button condition. Add a separate btnClearUpdateLock that calls
/api/updates/clear_lock -- never apply_force_update. The recovery
path never runs checkout/clean/reset --hard.
- CORE-1 (unconditional pre-cleanup): apply_force_update no longer
iterates .git/**/*.lock. Lock cleanup is exclusively the clear_lock
endpoint's job.
- Greptile P1: when a pull-lock error fires after a stash was pushed,
run git stash pop (with apply+drop fallback) so the user's local
modifications aren't stranded in the stash silently. v2 tests
cover the stashed and the clean cases.
- Greptile P2 (broad 'lock file' substring): tightened _GIT_LOCK_SIGNATURES
to specific git error strings ('index.lock': file exists, '.lock':
file exists, 'another git process seems to be running',
'unable to create .git/index.lock') so unrelated ref-transaction
lock-loss messages no longer trigger a lock-conflict response.
- Greptile P2 (redundant git_dir.exists / magic number): both gone as a
side effect of removing the apply_force_update cleanup loop entirely.
Refactor: rebased onto current upstream/master so PR is no longer
'behind' 130 commits.
Tests:
- 14 new v2 tests covering holder probe (true/false/NOGIL probe),
_try_remove_lock (refuse-on-held/success-on-unheld),
apply_clear_lock (success/refused), signature set parameter table
including false-positive class, apply_force_update no-touch
contract, pull-lock stash restore, pull-lock no-stash-when-clean.
- 2 v1 tests removed (they encoded the unsafe behavior).
- 2 v1 parametrize rows dropped (their positive cases relied on the
broad 'lock file' substring).
Closes#5687
- Detect lock file errors (lock_conflict flag) on fetch/status/pull
failures in _apply_update_inner (#1)
- Expose Force Update button on frontend when lock_conflict is
received (#3)
- Proactively clean stale lock files (>30s old) in apply_force_update
using rglob discovery instead of a hardcoded path list (#2)
- Add 10 tests: _is_git_lock_error parametrized unit tests,
_apply_update_inner lock path coverage, and apply_force_update
stale/recent lock behavior
- _repair_bare_custom_provider_model matches display-named providers via
_custom_provider_slug_from_name (custom:my-proxy matches 'My Proxy').
- _ordered_custom_provider_model_ids now handles dict keys, list strings,
and list dicts with id/model/name, aligned with api/config.py catalog.
- config_obj=None fallback uses get_config() instead of raw cfg.
- _read_profile_model_config returns profile config dict too, avoiding a
second YAML parse on hot display paths.
- Tests added for slug matching, list-form models, and get_config fallback.
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
Mirror the #5128 fast-path suffix repair when profile context is present
but requested_provider is omitted, which async delegation completion can
trigger via process_wakeup.
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
When session model_provider differs from profile model.provider, do not
expose profile_default_model for suffix repair. Fast path also requires
profile_provider to match requested_provider when profile_provider is set.
Addresses cross-provider mis-resolution (custom:other-proxy + bare suffix).
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
Refs #5127.
When _resolve_compatible_session_model_state takes the #1855 fast path with
a bare runtime model suffix and a slash-qualified profile default, repair to
the full qualified ID for custom/custom:* providers.
Also load profile model.default when the session already has model_provider
so wakeup turns can apply that repair.
Co-authored-by: b3nw <b3nw@users.noreply.github.com>
The SSE done handler previously used assistantText.slice(0, 100), which
captures the start of the live-stream accumulator (early tokens and interim
updates) rather than the final answer shown after tools settle.
Add _completionNotificationPreviewText() using anchor final-answer projection
and settled message content, with a cleaned live-stream fallback. Hoist
lastAsst so background completions on non-active panes still get preview text.
Update PWA notification regression tests.
Closes#5035
The update-check payload now carries a `dirty: bool` reflecting the
working-tree state vs HEAD, computed via the same `git diff-index
--quiet HEAD --` primitive already used by `_dirty_suffix()`. A dirty
install at-or-past the latest release tag used to silently report
"Up to date" with no remediation affordance; the Settings panel
can now read this flag to offer `apply_force_update` (which already
does `git checkout .` + `git reset --hard` to reset to a clean
latest).
The probe is conservatively clean on real errors (timeout, missing
git, fatal) so a transient probe failure never produces a false-
positive "local changes" alert.
Closes#4085
Resolve custom provider API keys from the matched config snapshot and pass them through session hydration plus streaming fallback context-length probes. This prevents authenticated /v1/models endpoints from falling back to the default 256K window and clobbering larger persisted session metadata.
When a custom/proxy provider serves models whose IDs share the same base
name across vendor prefixes (e.g. vendor_a/deepseek/deepseek-v4-pro vs
vendor_b/deepseek/deepseek-v4-pro), several normalization functions use
split('/').pop() (or split('/')[-1]) which discards all segments except
the last. This causes three user-facing symptoms: (1) clicking one
model selects a different colliding model, (2) configured-model badges
attach to the wrong dropdown entry, and (3) the model-chip label in the
composer bar is truncated to just the base model name.
Root cause: all three callers take only the last slash-segment instead
of stripping only the first (provider) segment and preserving the
remaining vendor hierarchy.
Fix 1 — _findModelInDropdown (static/ui.js): Move the exact string match
before the provider-aware normalized match. Previously, when all models
share the same provider ID (common with LLM proxy setups), the normalized
match returned whichever colliding option appeared first in DOM order,
even though an exact match existed.
Fix 2 — _normalizeConfiguredModelKey (static/ui.js) and _norm_model_id
(api/config.py): Replace split('/').pop() / split('/')[-1] with a first-
segment-only strip (regex on frontend, split('/',1) on backend), matching
the strategy already used by _findModelInDropdown's norm lambda. This
prevents multi-slash IDs from colliding in badge assignment and the
configured-entry dedup set. Additionally, strip colon-qualified provider
prefixes (e.g. custom:name/) before the slash strip so badge-key variants
like 'custom:llm-proxy/opencode_go/model' merge correctly with the bare
'opencode_go/model' in the configured section dedup.
Fix 3 — getModelLabel (static/ui.js) and _get_label_for_model
(api/config.py): Same split('/').pop() to first-segment-strip change so
the composer-bar model chip and backend label preserve vendor context
(e.g. shows 'opencode_go/deepseek-v4-pro' instead of 'deepseek-v4-pro').
Verification: 9 new regression tests (test_issue3360) covering exact-
match priority, multi-slash normalization, and backend/frontend parity.
Updated 1 existing test (test_norm_model_id_trailing_empty_guard) that
asserted the old split('/').pop() pattern. All 25 related tests pass.
AI Usage: Gemini (gemini-2.5-pro), via Antigravity IDE, pair-programmed.
(cherry picked from commit a454fecd2b3a83f7da34473c883be069171aeac9)
Two confirmed bugs in the thinking/reasoning display:
1. reasoningText was initialized once when the SSE stream opened and never
reset between turns. On the done event, the last assistant message
received the union of every turn's reasoning. Now reset at both turn
boundaries: tool (alongside existing liveReasoningText reset) and
interim_assistant (the other turn boundary where prior reasoning closes).
2. ui.js renderMessages preferred m.reasoning (which could be corrupted by
bug 1) over m.reasoning_content (the clean per-turn value from the
backend). The fallback now reads m.reasoning_content || m.reasoning.
Both fixes are needed: bug 2 alone cannot cover providers that stream
reasoning events without populating reasoning_content on the final API
message.
Updated test_streaming_race_fix.py to scope its reconnect-accumulator
guard to the _wireSSE preamble only, since turn-boundary resets inside
event listeners are intentional and correct.
9 new regression tests in test_issue2565_reasoning_accumulation.py.
Fixes#2713 — live assistant text can truncate at tool-call segment
boundaries during streaming.
Before _resetAssistantSegment() in the tool and interim_assistant SSE
handlers, synchronously flush any pending rAF render work so tokens that
arrived during the 66ms throttle window are written to the DOM before
assistantBody is cleared. Without this flush, the pending _doRender
callback fires after assistantBody is null and skips the write silently,
causing the tail of the pre-tool segment to disappear from the live view.
Implementation:
- Extract _flushPendingSegmentRender() helper (guarded by assistantBody
&& _renderPending) that cancels the pending rAF and synchronously
writes via smd/renderMd/esc — same cascade as _doRender.
- Call the helper from both the tool and interim_assistant handlers
before their respective _resetAssistantSegment() calls.
- Normal cases where the rAF has already fired are unaffected (guard
skips immediately).
Completed transcripts were never affected (renderMessages rebuilds from
the full assistantText accumulator on done).
Adds tests/test_issue2713_streaming_segment_flush.py with 11 static
analysis regression tests pinning the helper shape and call-site
ordering.
b3nw