…k router (lobehub#15146)

* ✨ feat(llm-generation-tracing): pre-allocate tracingId + recordFeedback router

Wire up the per-call feedback loop foundation.

1. **Pre-allocate tracingId (plan A2)**
   - `TracingOptions.tracingId?: string` — optional caller-supplied UUID.
   - `LLMGenerationTracingService.record` generates one via `randomUUID()`
     when the caller doesn't supply one, so the id is always known
     before DB insert.
   - `LlmGenerationTracingModel.record` accepts an optional `id` and
     forwards it to the insert (Drizzle still autogens when omitted).
   - `aiChat.outputJSON` allocates the id up-front, threads it through
     `tracing.tracingId`, and returns `{ data, tracingId }` so the
     client can wire feedback against the id even though
     `service.record` runs inside Next's `after()`.
   - `aiChatService.generateJSON` consumers (InputEditor, supervisor)
     unwrap the envelope.

2. **New `llmGenerationTracingRouter.recordFeedback`**
   - Scenario-agnostic feedback endpoint at `lambda.llmGenerationTracing`.
   - Validates `{ tracingId (uuid), signal (positive|negative|neutral),
     source, score?, data? }` and forwards to
     `LLMGenerationTracingService.recordFeedback`.

Follow-up issues already filed:
- LOBE-9488 — `@lobehub/editor` AutoCompletePlugin needs
  `onAccept`/`onReject`/`onCancel` callbacks before the client side can
  capture Tab/Esc/keep-typing signals against the returned tracingId.
- LOBE-9489 — session-level signal modeling (multi-suggestion typing
  sessions) — deferred until per-row feedback data lands.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* 🐛 fix(llm-generation-tracing): surface feedback write failures instead of silent ok

The recordFeedback mutation used to always return `{ ok: true }` even when
the underlying write was silently dropped — `LLMGenerationTracingService`
swallowed both DB-init/update throws and the no-op case where the WHERE
clause (id + userId) matched zero rows. Callers couldn't tell
"persisted" from "lost", which would skew tracing-feedback metrics and
prevent reasoned retry/error handling.

Fix:

- `LlmGenerationTracingModel.updateFeedback` now returns
  `{ updated: boolean }` (via `.returning({ id })`), so the caller knows
  whether the WHERE clause actually matched a row.
- `LLMGenerationTracingService.recordFeedback` throws a typed
  `LLMGenerationFeedbackError` with `kind: 'not_found' | 'db_failure'`
  instead of swallowing — stops logging-only behaviour for DB errors and
  promotes the 0-rows case to an explicit signal.
- `llmGenerationTracingRouter.recordFeedback` catches that error and
  translates to `TRPCError({ code: 'NOT_FOUND' })` for stale-id and
  `INTERNAL_SERVER_ERROR` for DB outages — `{ ok: true }` only flows
  back when a row was actually patched.

Tests:
- Model: assert `{ updated: true/false }` for happy / cross-user / missing-id
- Service: assert throws on both not_found scenarios
- Router: assert TRPCError code translation for both error kinds

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* ✨ feat(input-completion): wire Tab/Esc/typing feedback to recordFeedback

- bump @lobehub/editor to ^4.12.0 for AutoComplete onSuggestion{Accepted,Rejected}
- add llmGenerationTracingService wrapping lambda.llmGenerationTracing.recordFeedback
- InputEditor: map suggestionId→tracingId, fire positive on accept, negative on
  esc, neutral on typing/cursor-move/blur/other; recode IME-driven escape as
  neutral/autocomplete_ime so CJK input doesn't poison the signal

Closes LOBE-9488

* ♻️ refactor(input-completion): fold recordTracingFeedback into aiChatService

Single trpc mutation didn't warrant a dedicated service file; aiChatService
already owns the paired `outputJSON` call that mints the tracingId, so
recordTracingFeedback belongs alongside it.

* 💄 style(llm-generation-tracing): tag task-handoff scenario + prompt version (lobehub#15191)

* 💄 style(QueueTray): use borderless variant for queued file preview

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* ✨ feat(llm-generation-tracing): tag task-handoff scenario + prompt version

Task topic handoff was tracing as scenario=unknown / promptVersion=v0 because the
generateObject call only set metadata.trigger and that trigger isn't in the
registry. Add a TaskHandoff scenario const, version the prompt next to its
definition, and pass tracing options explicitly at the call site (mirroring
followUpAction).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* 🐛 fix(llm-generation-tracing): validate caller-supplied tracingId as UUID

The `outputJSON` route echoed `tracing.tracingId` back to clients without
checking the shape. Because the surrounding `tracing` record is free-form,
a malformed value passed request validation, then failed DB insertion on
the uuid PK and was later rejected by `recordFeedback` (`z.string().uuid()`),
so callers could receive a tracingId unusable for the feedback flow.

Tighten `StructureOutputSchema.tracing` to a `z.object({ tracingId: uuid }).catchall(unknown)`
so the validation happens at the request boundary; the route can then drop
the redundant `typeof === 'string'` guard.

---------

Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>

# 🚀 LobeHub Release (20260528)

**Release Date:** May 28, 2026  
**Since v2.2.0:** 220 merged PRs · 15 contributors

> This cycle brings heterogeneous "platform agents" you can dispatch to
local or remote devices, a rebuilt onboarding flow, document-centric
chat, and a unified model-runtime error model — with new DeepSeek V4 and
Gemini 3.5 Flash support along the way.

---

## ✨ Highlights

- **More Hetero Agents (OpenClaw / Hermes)** — Create heterogeneous
agents and dispatch them to local or remote devices through the device
gateway, with an execution-target switcher in the composer and
persistent CLI sessions. (#15065, #15179, #15022)
- **iMessage on Desktop** — New iMessage setup and bridge on desktop,
plus bot attachments across every platform. (#15228, #15227, #15029)
- **Skills in the Composer** — Drag skill chips into chat, trigger
installed skills from the slash menu mid-line, and surface project-level
skills in the homogeneous agent runtime. (#15095, #15061, #15110)
- **New Models** — DeepSeek V4 Flash/Pro and Gemini 3.5 Flash across
providers, with thinking params for structured output and chat cost
estimates. (#15031, #15001, #15051, #14876)
- **Agent Runtime Observability** — OpenTelemetry GenAI semantic
conventions plus per-call generation tracing. (#15123, #15124)

---

## 🤖 Agents & Heterogeneous Runtime

- **Platform agent creation** — OpenClaw/Hermes creation UI, device
guard, and remote dispatch backend. (#15065)
- **Execution-target switcher** — Pick local vs remote execution
directly in the composer; device-selection UX with actionable guidance.
(#15179, #15111)
- **CLI hetero dispatch** — OpenClaw/Hermes dispatch with persistent
sessions and a notify protocol. (#15022)
- **Gateway snapshot as source of truth** — Consume the gateway
`uiMessages` snapshot at step boundaries to keep chat state consistent.
(#15153, #15152)
- **Client sub-agent as a normal tool call** — Simplifies the sub-agent
execution path. (#15281)
- **Hermes agent chain** — Implements the Hermes agent chain logic.
(#15189)
- **Device registry** — TRPC endpoints to register, list, update, and
remove devices. (#15299)
- **Desktop device routing** — Route gateway agent runs through `lh
hetero exec`; restore `userId` in gateway dispatch and gate local-system
by execution target. (#15132, #15232)
- **Agent signals** — Anchor agent-signal receipts to messages and
isolate memory-agent messages into a child thread. (#14969, #14921)

---

## 🚀 Onboarding

- **Simplified first screen** — Defer topic creation to first send.
(#15090)
- **Market Agent Picker** — Added as a classic onboarding step, with
template prefetch. (#14980, #15041)
- **Welcome guidance** — Show agent welcome guidance on first run.
(#15098)
- **Mobile** — Adapt agent onboarding UI and restore Classic-step
padding on mobile. (#15019, #15032)
- **Discovery** — Streamline discovery to a single profession question.
(#14987)
- **Analytics** — Track onboarding step events and create-agent modal
source. (#15133, #15028)

---

## 📄 Documents, Pages & Knowledge

- **Thread chat in preview** — Embed thread chat in the document preview
portal. (#15216)
- **Non-markdown rendering** — Render non-markdown docs as a read-only
highlight. (#15272)
- **Multi-select** — Multi-select delete in the document tree. (#15125)
- **Page-agent streaming** — Preview `initPage` streaming arguments.
(#15039)
- **Per-agent topics** — Per-agent topic management page. (#15207)
- **Server-side category** — Derive document category server-side and
drop frontend predicates. (#15076)

---

## 🧩 Skills & Tools

- **Drag skill chips** — Drag skills into chat input and register
agent-document skills. (#15095)
- **Slash menu** — Installed skills appear in the slash menu with a
mid-line trigger. (#15061)
- **Project skills** — Recognize project-level skills in the homogeneous
agent runtime and surface them regardless of active device. (#15110,
#15177)
- **VFS archiving** — Archive oversized tool results to VFS instead of
truncating. (#15074)
- **@localfile mentions** — Drag folders into chat input as `@localFile`
mentions on desktop. (#15071)

---

## 🧠 Model Runtime & Providers

- **Error spec registry** — Unify error codes into a spec + pattern
registry, split `ProviderBizError` into finer codes, classify Cloud-only
codes via a tier digit, and add `DatabasePersistError`. (#15262, #15286,
#15278, #15279)
- **New models** — DeepSeek V4 Flash/Pro (opencode-go) and Gemini 3.5
Flash; DeepSeek V4 Pro on SiliconCloud. (#15031, #15001, #15017, #15267)
- **Structured output** — Thinking params for structured output, Bedrock
structured generation, and DeepSeek `generateObject` tool choice.
(#15051, #15174, #15054)
- **Cost** — Chat cost estimate support; preserve usage cost in custom
streams. (#14876, #15218)

---

## 💬 Chat & User Experience

- **Follow-up chips** — Extend follow-up chip suggestions to general
chat with scene-specific model config. (#15101, #14797)
- **Input drafts** — Persist unsent input drafts across tab switches and
prevent repeated draft restore. (#14992, #15024)
- **Command menu** — Order topic/message search by recency and promote
inline type filters. (#15094, #14986)
- **Zoom HUD** — Show a zoom-level HUD on Cmd +/− and Cmd 0. (#15294)
- **Copy** — Unescape markdown escapes when copying user messages.
(#15253)

---

## 🖥️ Desktop

- **App Nap fix** — Prevent App Nap from dropping the gateway WebSocket
during display sleep. (#14994)
- **File preview** — Preview `.cjs`/`.mjs`/no-extension files instead of
binary fallback and expand `~` when opening local files. (#15168,
#15284)
- **Cross-platform settings** — Open settings via main-window navigation
on Windows/Linux and restore the route after an update restart. (#15036,
#14922)
- **Token refresh** — Prevent frequent logout from token-refresh
retries. (#14928)

---

## 📊 Observability

- **OTel GenAI** — Instrument Agent Runtime with OpenTelemetry GenAI
semantic conventions. (#15123)
- **Generation tracing** — Per-call `llm_generation_tracing` with a
pre-allocated tracingId and recordFeedback router. (#15124, #15146)
- **Error classification** — Persist `ERROR_CODE_SPECS` classification
on operation errors. (#15273)

---

## 🗃️ Database Migrations

- **Batch migrations** — Topic usage stats, push tokens,
`tasks.editor_data`, and document shares. (#15280)
- **Tracing & eval tables** — Add `llm_generation_tracing` and agent
eval experiment tables. (#15126)

> Self-hosted operators should run the database migration (`pnpm
db:migrate`, or restart with auto-migrate enabled) after upgrading. The
changes are additive and backwards-compatible.

---

## 🔒 Security & Reliability

- **Security:** Remove the `getPlaintextCred` tool to prevent plaintext
credential exposure. (#14998)
- **Security:** Prompt account selection for Google OAuth and add
`prompt=consent` to the OIDC authorization URL to fix missing refresh
tokens. (#15234, #15010)
- **Reliability:** Preserve streamed content across a mid-stream cancel.
(#15173)
- **Reliability:** Bound the Redis command timeout and configure the
Anthropic client timeout. (#15091, #15042)
- **Reliability:** Prevent infinite recursion in the assistant chain.
(#15288)

---

## 👥 Contributors

Huge thanks to **15 contributors** who shipped **220 merged PRs** this
cycle.

@AnotiaWang · @sxjeru · @algojogacor · @hardy-one · @arvinxx · @Innei ·
@tjx666 · @lijian · @AmAzing129 · @rdmclin2 · @neko · @cy948 ·
@CanisMinor · @sudongyuer · @rivertwilight

Plus @lobehubbot and renovate[bot] for maintenance.

---

**Full Changelog**: v2.2.0...release/weekly-20260528