* fix(auxiliary): route direct-create aux callers through call_llm (#35566)
Five callers (kanban_decompose, kanban_specify, profile_describer, and
goals.py's judge + draft-contract) built raw clients via
get_text_auxiliary_client() and passed extra_body=get_auxiliary_extra_body()
— which only returns Nous portal tags and ignores
auxiliary.<task>.extra_body from config.yaml entirely. That was the
remaining half of #35566 after the call_llm path was fixed.
Routing them through call_llm(task=...) gives each caller the full
auxiliary contract for free: task extra_body, the reasoning_effort
shorthand, transient retries, provider-profile projection, and fallback
chains. goal_judge gains a DEFAULT_CONFIG block (it had none — its
provider/model overrides silently didn't exist as documented keys).
get_auxiliary_extra_body() now has zero non-test callers; kept for
plugin back-compat.
Fixes#35566.
* test: migrate kanban dashboard + CLI specify mocks to call_llm
Two more consumers of specify_task mocked the old
get_text_auxiliary_client symbol (missed in the first sibling sweep —
they live outside tests/hermes_cli's kanban files): the dashboard
plugin's /specify endpoint tests and the /kanban slash-command E2E.
Same migration as the rest: mock call_llm at the source, no-provider
now surfaces via the LLM-error branch.
Ollama's /v1/chat/completions silently ignores extra_body.think (it only
honours it on /api/chat — ollama/ollama#14820), so agent.reasoning_effort:
none never actually disabled thinking on OpenAI-compatible Ollama routes.
Emit the top-level reasoning_effort='none' field (which Ollama respects)
alongside think=False (kept for proxies and the native /api/chat path).
The PR's second half (propagating reasoning_config to the background-review
fork) already landed on main via agent/background_review.py, so only the
provider-profile change is salvaged here, resolved onto the current
GLM/effort-aware profile.
Salvaged from PR #29820 by @Epoxidex.
Five tests for the salvaged #37217 Bug B fix: vendor-field passthrough,
reasoning-key + private-key exclusion, merge-over-existing (fast-mode
speed), no-extra_body regression guard, reasoning-only adds nothing.
Live probes against api.anthropic.com informed the exclusion design:
Anthropic strictly validates the request body (unknown keys 400 with
'Extra inputs are not permitted'), so the passthrough forwards only
caller-configured fields and never the OpenAI-shaped reasoning dict
(translated natively) or _-private plumbing keys.
Two related bugs in _AnthropicCompletionsAdapter.create() in
agent/auxiliary_client.py silently discard caller-supplied
reasoning_config and extra_body on the Anthropic-Messages
auxiliary-protocol path:
* Bug A: reasoning_config=None was hardcoded at L1000, so the
reasoning_config parameter on build_anthropic_kwargs was
unreachable for any auxiliary task. The main agent path
(agent/transports/anthropic.py) already reads
reasoning_config from caller params; this PR aligns the
auxiliary adapter with the same pattern.
* Bug B: create(**kwargs) accepts an OpenAI-style kwargs
payload from the caller but only forwards a hand-picked
subset to self._client.messages.create(). Any caller-supplied
extra_body (e.g. thinking control, metadata, service_tier,
vendor-specific fields) was dropped on the floor. The
codex/responses transport in the same file already merges
extra_body; the Anthropic branch is the gap.
This unlocks the caller-supplied extra_body path so auxiliary
callers can set per-vendor request fields (including
thinking: {type: "disabled"} for Anthropic-compatible vendors
that require an explicit disable on the wire), and lets the
reasoning_config kwarg flow into build_anthropic_kwargs like the
main agent does. Both changes are backward-compatible for
callers that don't pass the affected kwargs.
Affected providers (all routed through _AnthropicCompletionsAdapter
via _maybe_wrap_anthropic): anthropic (native), minimax /
minimax-cn, kimi-coding / kimi-coding-cn, z.ai / GLM, and any
custom /anthropic-suffixed endpoint. See PR description for
related issues (#35566, #7209, #16533, #32813, #29248).
Catalog entries now follow the same supply-chain rules as pyproject
dependencies:
- n8n: install.ref main -> full commit SHA 7a9ae007 (2026-05-23,
branches/tags can be moved by the upstream owner; SHAs cannot)
- new contract test: every shipped manifest must pin exactly —
git installs need a 40-char SHA, uvx/npx-style launchers need
pkg==X / pkg@X with a digit-leading version (rejects bare names,
ranges, and npm dist-tags like @latest)
- module docstring documents the pin policy (exact version, 2-week
cooldown)
unreal-engine and linear are http transports (server runs elsewhere)
so there is nothing to pin at the transport layer.
Verified: unpinning blender-mcp in the manifest makes the contract
test fail with a named diagnostic; restoring the pin passes.
MCP catalog entries follow the same supply-chain rules as pyproject
dependencies: exact version pin, and the pinned release must be at
least 2 weeks old. blender-mcp 1.6.4 released 2026-06-11 (~5 weeks
old, also the latest release). uvx now resolves the exact version
instead of latest-at-launch.
Adds optional-mcps/blender (ahujasid/blender-mcp, stdio via uvx). The
server advertises 22 tools; 18 front optional asset services with no
upstream trim mechanism, so tools.default_enabled pins the install to
the core surface (scene/object info, viewport screenshot, code exec)
and the rest stay opt-in through 'hermes mcp configure blender'.
Manifests can now declare transport.env (static, non-secret subprocess
env vars), parsed/validated in _parse_manifest and written by
_build_server_config — used here to ship DISABLE_TELEMETRY=true per
the no-telemetry-without-opt-in policy. Runtime already honored
per-server env; manifests just couldn't declare it.
With the TTFB watchdog now scaled (not disabled) for large requests on
main, the hard ceiling is a backstop against mid-stream wedges, not the
primary stall detector. The original 600s default clamped BELOW the
intentional 1200s stale floor for >100k-token requests, partially
reverting the floor that keeps healthy gateway-scale payloads alive.
Raise the default to 1500s so the ceiling only catches unbounded
growth, never healthy slow turns.
A large Codex request (estimated context >= 10k tokens) disables the no-byte
TTFB watchdog on purpose, and openai_codex_stale_timeout_floor *raises* the
stale timeout (up to 1200s at >100k tokens) so healthy gateway-scale payloads
aren't aborted mid-prefill. When the backend genuinely stalls — no first byte
AND no events, exactly the #64507 symptom — the request is only reclaimed at
that high stale floor, so the session can hang for 13+ minutes with an idle
slash_worker and no ended_at/end_reason while Desktop still shows it as active.
Add a flat, finite hard ceiling on total openai-codex request time that always
applies (min() of the computed stale timeout and the ceiling) regardless of the
TTFB-disable / stale-floor interaction. A stalled large request is now killed
at the ceiling and the retry loop / visible failure path takes over instead of
hanging indefinitely. Tunable via HERMES_CODEX_HARD_TIMEOUT_SECONDS (default
600s; 0 disables the ceiling to restore pre-fix behavior).
Closes#64507
TRUNCATE checkpoint every 50 writes causes B-tree corruption on large
databases (65K+ pages) due to the exclusive-lock I/O pressure from
checkpointing thousands of frames at once.
Switch periodic _try_wal_checkpoint() to PASSIVE mode which does not
require an exclusive lock and cannot corrupt pages under I/O pressure.
Keep TRUNCATE in close() and pre-VACUUM paths where it is safe
(infrequent, controlled conditions).
Also replace silent `except Exception: pass` with logged warnings for
checkpoint failures so operators can detect early corruption signals.
Fixes#45383
_summarize_tool_result() built the web_extract summary straight from the
first `urls` entry. When web_search results are forwarded into web_extract
(a common chain), that entry is a dict ({"url"/"href": ...}) rather than a
URL string. With 2+ URLs the `url_desc += " (+N more)"` step then raised
`TypeError: unsupported operand type(s) for +=: 'dict' and 'str'`, aborting
the pre-compression pruning pass; with a single URL the raw dict repr
leaked into the summary text.
Unwrap the URL from dict entries before use — mirroring the existing
web_extract handling in agent/display.py (`_display_url`),
tools/web_tools.py (`_web_extract_url`) and acp_adapter/tools.py
(`build_tool_title`) — so `url_desc` is always a string and the
concatenation is always str + str.
Adds regression tests covering multiple/single dict URLs, the href key,
malformed dicts, and the plain-string path.
test_normalize_codex_response_salvage_is_xai_scoped broke on main when
two same-day merges crossed: #64764 (#64434 — trust response.status for
reasoning-only turns on UNRECOGNIZED Responses backends) changed what a
bare _normalize_codex_response(response) call returns for
status='completed' reasoning-only output (now 'stop'), while #64768
added this test calling with no issuer_kind and expecting 'incomplete'.
The test's intent is that the xAI reasoning-channel salvage does not
leak into other special-cased backends — pin issuer_kind='codex_backend'
so it exercises exactly that (same pattern as
test_normalize_codex_response_treats_summary_only_reasoning_as_incomplete,
which was already pinned for #64434).
Add merge_existing to save_config (default False for full-document callers
like the dashboard YAML editor) and route partial writes through
_merge_partial_save. _persist_migration writes the full migrated dict
directly so deleted keys are not resurrected from the on-disk file.
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 #64434 change makes unrecognized issuers trust
response.status='completed' for reasoning-only turns, so this sibling
test (which exercised the old default-path behavior) now pins the Codex
backend explicitly — the surface where reasoning-only still means
'still thinking'.
Follow-up to the salvaged #64449: the status-trusting branch flipped
github_responses to 'stop' alongside unknown relays. Copilot fronts the
same OpenAI model family as codex_backend and shows the same
reasoning-only 'still thinking' degeneration, so it stays on the
continuation path. Only unrecognized (other:*) backends trust
response.status='completed' as terminal.
The aggregator is MoA's acting model, but the main loop's reasoning
gates key off the virtual moa://local identity and never fire — so with
no per-slot reasoning_effort the aggregator silently ran at the backend
default, ignoring the user's reasoning config entirely (#64187).
New _aggregator_reasoning_config(): slot value > full acting-model
resolution via the shared chokepoint (agent.reasoning_overrides for the
slot's model > global agent.reasoning_effort; YAML False stays
'disabled'). Applied to both aggregator call sites (acting turn +
one-shot /moa synthesis).
Reference advisors intentionally keep slot-or-default: inheriting a
global xhigh into every advisor fan-out would silently multiply cost.
Fixes#64187.
The remote model catalog (website/static/api/model-catalog.json) now labels
exactly one entry per provider block with "default": true — z-ai/glm-5.2 for
both OpenRouter and Nous Portal. That labeled entry is the model Hermes
silently lands on when the user never picked one, and it can be rotated by
editing the manifest alone: no release needed.
- model_catalog.py: get_default_model_from_cache() reads the label from the
in-process/disk cache only — never triggers a network fetch, so hot
resolution paths (agent build, gateway session setup) stay network-free.
- models.py: get_preferred_silent_default_model() resolves catalog label
first, PREFERRED_SILENT_DEFAULT_MODEL constant second (offline/fresh
install). _PROVIDER_SILENT_DEFAULT_OVERRIDES dict replaced by
_SILENT_DEFAULT_PROVIDERS routing through the shared resolver.
fetch_openrouter_models() preserves the "default" badge through live
/v1/models refreshes so the picker shows it.
- scripts/build_model_catalog.py: generator emits the default label so
regeneration can't drop it.
- website/docs/reference/model-catalog.md: schema documents the new field.
- Salvaged from PR #61141 (@HumphreySun98): bare-provider /model switches
(/model nous) route through the cost-safe default instead of curated
entry [0].
- tests: catalog-label precedence, constant fallback, stale-label fallback,
cache-only (no network) guarantee, and a shipped-manifest contract test
pinning the labeled entry to PREFERRED_SILENT_DEFAULT_MODEL.
E2E (temp HERMES_HOME): fresh-install constant fallback, shipped-manifest
label read, release-free rotation (relabeled cache -> new default across
models.py, tui_gateway, and gateway empty-model paths) all verified.
`detect_static_provider_for_model` handles a bare provider name typed as a
model (e.g. `/model nous`) by returning `_PROVIDER_MODELS[provider][0]` — the
first curated entry. For metered aggregators whose curated list is ordered
most-capable-first (Nous Portal), entry [0] is the priciest flagship, so
`/model nous` silently switched to it.
This is exactly the billing footgun `_PROVIDER_SILENT_DEFAULT_OVERRIDES` /
`get_default_model_for_provider` exist to prevent (per their docstring, a
missing model "escalated to Opus and billed 863 requests before the user
noticed"). The non-interactive fallback already routes through that cost-safe
helper; the interactive `/model <provider>` path did not.
Route this path through `get_default_model_for_provider` too. Providers
without a silent-default override are unchanged (the helper returns
`models[0]`), so only overridden providers (currently `nous`) change — from
the flagship to the low-cost default.
Adds regression tests: `/model nous` resolves to the cost-safe default, and
non-overridden providers still resolve to their first catalog model.
Consolidation follow-up on top of @liuwei666888's compressor fix (#52431)
and @Frowtek's ACP render guard (#62588) — completes the class fix across
every display-layer consumer of tool-call arguments:
- agent/context_compressor.py: wrap _summarize_tool_result in a
never-raises backstop (same pattern as get_cute_tool_message and the
ACP guard). The per-branch _str_arg coercions keep summaries
informative; the wrapper guarantees compression can never crash-loop
on a summary branch we didn't anticipate. Also reject non-dict parsed
args (a JSON list/scalar would crash every args.get call).
- agent/display.py: build_tool_preview's process branch sliced
session_id/data without coercion — crashed build_tool_preview and
build_tool_label on the live tool-progress callback (probed live:
TypeError 'int' object is not subscriptable). Coerce like the
sibling branches.
- tests: backstop fuzz matrix (16 tools x 6 hostile value shapes),
fallback-shape contracts, display process-preview regressions,
non-dict args guard.
build_tool_start renders every ACP (Zed) tool call — on the live tool-progress
callback (acp_adapter/events.py) and during session history replay
(acp_adapter/server.py). It called build_tool_title and extract_locations
directly, so a model that emits a malformed argument crashed the render:
- terminal `command` as null/number -> TypeError (len() in build_tool_title)
- delegate_task `goal` as a number -> TypeError (len())
- read_file `path` as a non-string -> pydantic ValidationError building a
ToolCallLocation
A live crash breaks the tool-call event; a persisted one breaks history replay
on every resume of that session. The sibling CLI label builder
get_cute_tool_message was already wrapped for exactly this reason
(agent/display.py: "display must never abort a turn").
Wrap build_tool_start the same way: on any builder failure, fall back to a
minimal, valid start event (tool name as title, resolved kind). The happy path
is unchanged.
Adds tests for the non-string command, path, and goal cases.
When LLMs return non-string parameter values (e.g. bool, int) in tool call
arguments, _summarize_tool_result() crashes with TypeError because it calls
len(), .count(), or slicing directly on args.get() return values.
This causes an infinite crash loop in the TUI — context compression
triggers on session resume, which crashes, which restarts, which triggers
compression again.
Add _str_arg() helper that coerces any value to str, and use it in all 5
vulnerable call sites:
- terminal: len(cmd)
- write_file: content.count()
- delegate_task: len(goal)
- execute_code: len(code)
- vision_analyze: question[:50]
* test: deflake async-delegation interrupt test under CI load
test_interrupt_all_signals_running_children failed twice on a loaded CI
worker: the blocker's ev.wait(timeout=5) expired before interrupt_all()
ran, the record finalized on its own, and interrupt_all() found nothing
running (n == 0, interrupted count 0 — the exact CI assertion failure,
reproduced locally by inserting a 5.6s dispatch->interrupt gap).
Raise the internal safety timeouts from 5s to 60s across the file's
gated runners — they exist only as runaway guards (every test releases
its gate explicitly); the pytest-level timeout is the real backstop.
Same flake class as the removed test_crashed_runner_produces_error_
completion (#64431).
* test: fix cross-test completion-event leak (second flake mechanism)
The first CI failure was the 5s guard-timeout race; the rerun exposed a
SECOND mechanism with a different signature: 'completed' == 'interrupted'.
Prior tests (e.g. test_dispatch_rejected_at_capacity) release their gate
and return immediately, but their workers finalize asynchronously — on a
loaded runner the teardown drain races the in-flight _finalize, and the
straggler 'completed' events leak into the NEXT test's queue, where
_drain_one() picks one up instead of the interrupt event. Reproduced
locally: gate release + immediate drain + 0.15s finalize delay leaked 2
events.
Fixes:
- teardown waits (bounded 2s) for active workers to finalize before
draining, so events land in the owning test
- the interrupt test matches its OWN delegation_id via _drain_for()
instead of taking whatever event arrives first
The 0.14 upgrade (#63970) dropped the @assistant-ui/store override, so the
whole cluster de-hoisted from root node_modules into apps/desktop/node_modules
under a single shared tap@0.9.3. The test only looked at the root hoist path
(node_modules/@assistant-ui/tap), which no longer exists, and failed on main.
Resolve tap wherever npm places it (root or workspace-nested), and assert a
single shared version across all install sites — strengthening the invariant to
also catch a split tap install, not just split declared ranges.
The assistant-ui upgrade lockfile omitted standalone @esbuild/* packages.
CI runs npm ci --ignore-scripts, so esbuild's postinstall never runs and
desktop/ui-tui check (build:ink, bundle-electron-main) fail without them.
Graft the 26 platform packages from main's lockfile.
@assistant-ui/react 0.14 makes respondToApproval required on
ToolCallMessagePartProps; the settledClarifyProps helper still lacked it
after the upgrade cherry-picks, so tsc failed on clarify-tool.test.tsx.
a reader of this subclass can't recover from hermes code alone that the metadata.isOptimistic flag drives core's off-branch eviction and export() omission, so a future core change to it would silently break placeholder cleanup. flagged in the upgrade review.
lock the four behaviors the built-in normalizeMathDelimiters/escapeCurrencyDollars introduce over the deleted custom helpers: $$<digit>$$ display math stays intact, double-backslash brackets and [/math]/[/inline] tag pairs rewrite to dollar delimiters, and currency dollars in prose are escaped. the existing preprocessMarkdown suite had no math cases.
the parseIncompleteMarkdown comment implied the reveal frontier is repaired; repair runs on the full accumulated text, so reword it to say that. drop the now-dead "multiple surfaces render the same content" clause from the block-cache comment (the smooth and defer wrappers that caused it were removed), and trim the math-preprocess comment to the load-bearing prose-only constraint.
@assistant-ui/store renamed its index-out-of-bounds throw from tapClientLookup/tapClientResource to useClientLookup in the 0.14 upgrade, so the boundary's /tapClient.../ filter stopped matching and re-threw the transient session-switch and reconnect race to root, blanking the app. broaden the regex to accept the new prefix (keeping the old one for older store versions) and point the test at the real message so it exercises the live path instead of the dead string.
delete the custom rewriteLatexBracketDelimiters and escapeCurrencyDollars
implementations from markdown-preprocess.ts (~40 lines). the built-in
exports from @assistant-ui/react-streamdown 0.3.4 are strict
improvements:
- normalizeMathDelimiters combines rewriteLatexBracketDelimiters (now
handles double backslashes and trims body whitespace) with
rewriteCustomMathTags (handles [/math]...[/math] and
[/inline]...[/inline] tags that some models emit — new capability
HA didn't have before)
- escapeCurrencyDollars excludes $ as a preceding character, so
display math $$5 is no longer incorrectly escaped (bugfix)
the call site in preprocessMarkdown changes from
rewriteLatexBracketDelimiters(escapeCurrencyDollars(part)) to
normalizeMathDelimiters(escapeCurrencyDollars(part)).
verified: tsc 0 errors, eslint clean, all 16 preprocessMarkdown tests
pass (including currency dollar escaping), vitest 0 new failures,
manual verification of currency amounts, LaTeX bracket delimiters,
display math, and dollar signs inside code blocks.
delete lib/remend-tail.ts (108 lines) and lib/remend-tail.test.ts (105
lines). the tailBoundedRemend export from @assistant-ui/react-streamdown
0.3.4 is algorithmically identical — same findRemendWindowStart boundary
scan, same fence/math tracking, same slice-and-repair strategy. the only
differences are improvements: the built-in handles \r (CR) in line
endings for Windows compatibility, and accepts an optional RemendOptions
parameter passed through to remend.
the import in markdown-text.tsx moves from @/lib/remend-tail to
@assistant-ui/react-streamdown. the call site
(preprocessWithTailRepair) is unchanged.
verified: tsc 0 errors, eslint clean, vitest 0 new failures (15
pre-existing, 786 passing — 6 fewer than before because the deleted
remend-tail.test.ts had 6 cases), manual verification of incomplete
markdown repair during streaming.
delete SmoothStreamingText, DeferStreamingText, and useSmoothReveal
(~174 lines) from markdown-text.tsx. the built-in defer and smooth
props on StreamdownTextPrimitive now handle the same work:
- defer: routes streaming text through useDeferredValue so markdown
re-parsing runs at lower priority (typing/scrolling stay responsive)
- smooth: typewriter-style reveal via useSmooth with SmoothOptions
{ drainMs: 500, maxCharsPerFrame: 30, minCommitMs: 33 }, matching
the old useSmoothReveal constants exactly
MarkdownTextContent (reasoning text) gets both defer and smooth.
MarkdownText (assistant text) gets defer only, matching the previous
behavior where text messages had no typewriter effect.
the internal pipeline order changes from smooth → defer → preprocess
to preprocess → smooth → defer (the built-in primitive runs preprocess
first). this is functionally equivalent: the tail-bounded remend repair
runs once on the full text instead of per revealed prefix, and the
smooth reveal operates on already-repaired markdown. end result is
identical.
verified: tsc 0 errors, eslint clean, vitest 0 new failures (15
pre-existing, 792 passing), manual verification of 6 streaming
scenarios (defer, smooth reveal, typing-while-streaming, code blocks,
math, long text performance).
bumps @assistant-ui/react from ^0.12.28 to ^0.14.23 and
@assistant-ui/react-streamdown from ^0.1.11 to ^0.3.4. this crosses
two minor bumps on each package and unlocks the built-in defer, smooth,
and tail-bounded remend primitives for PR 2.
breaking change from core 0.2.x: MessageRepository.appendOptimisticMessage
was removed (assistant-ui#4162). inline the three steps it did (generateId
+ fromThreadMessageLike + addOrUpdateMessage) in
incremental-external-store-runtime.ts, and set metadata.isOptimistic so
the new off-branch eviction logic cleans up the placeholder correctly.
fromThreadMessageLike and generateId graduated to the public API in
0.14.22 (assistant-ui#4414), so they now import from @assistant-ui/react
instead of @assistant-ui/core/internal. ExportedMessageRepository in the
test file moves to the public import for the same reason. the remaining
internal imports (AssistantRuntimeImpl, BaseAssistantRuntimeCore,
ExternalStoreThreadListRuntimeCore, ExternalStoreThreadRuntimeCore,
hasUpcomingMessage) are runtime construction internals with no public
equivalent and stay on @assistant-ui/core/internal.
the @assistant-ui/store npm override is removed: all transitive ranges
now resolve to 0.2.18 without it.
verified: tsc --noEmit passes, vitest shows zero new failures (15
pre-existing, 792 passing, identical to baseline before the upgrade).
The chat_completions path raises EmptyStreamError when a stream yields
no chunks and no finish_reason (#64420). _call_anthropic() had no
equivalent, and the eventless failure surfaces differently by client:
- Real Anthropic SDK: no message_start means no final-message snapshot,
so get_final_message() raises a bare AssertionError — not in the
retry loop's transient set, so it burned no retries and surfaced raw.
- OpenAI-compat shims: may fabricate a contentless Message with no
stop_reason (or return None), flowing out as a 'successful' empty turn.
Track whether any stream event arrived and normalize both shapes to
EmptyStreamError, giving the anthropic path the same transient retry
budget in the shared _call() retry loop. A real completed response
always carries a stop_reason, so the guard cannot fire on legitimate
turns (including eventless mocks with stop_reason='end_turn').
Follow-up to #64420.
An exhausted EmptyStreamError previously reported 'Connection to provider
failed' — misleading, since the connection succeeded (stream opened) and
the provider simply sent nothing. Add a dedicated third branch so users
debugging a misconfigured endpoint aren't sent chasing network issues.
Follow-up to #64420.
Three-layer companion to the salvaged CLI drain-ownership fix (#64240):
1. restore_undelivered_completions stamps restored=True (in-memory only)
on every durable completion re-enqueued at process start.
2. drain_notifications' legacy unfiltered branch re-queues restored
events instead of consuming them — a fresh process can no longer
adopt a dead session's delegation results (#64484). Same-process
keyless events keep the legacy behavior.
3. delegate_tool's async dispatch now falls back to the parent agent's
durable session_id when the approval-context key resolves empty (the
CLI case), so the CLI's new positive-ownership drain can actually
claim its own completions instead of failing closed on ''.
