…15179)

* ✨ feat(hetero-agent): add execution target switcher in composer

Add a chip in the chat composer toolbar that lets users pick where a
heterogeneous agent (claude-code / codex) executes: on this desktop, in
a cloud sandbox, or on an `lh connect` remote device. Persists the
choice via a new `agencyConfig.executionTarget` field paired with the
existing `boundDeviceId`. Server dispatch wiring will land separately.

* 🐛 fix(hetero-agent): mount execution target switcher in hetero composer

The hetero `ChatInput` replaces `RuntimeConfig` with `WorkingDirectoryBar`
via `runtimeConfigSlot`, so the new chip added in the previous commit
was never reached for hetero agents. Mount `HeteroDeviceSwitcher` in
`WorkingDirectoryBar` directly (both desktop and web branches).

* 💄 style(hetero-agent): polish execution target popover

- Drop uppercase + letter-spacing from section titles for normal sentence case
- Add a green status dot next to "Online" on device rows
- Rename "Remote devices (lh connect)" to "Other devices" with a clarifying
  subtitle so it covers both desktop-app and `lh connect` machines

* 💄 style(hetero-agent): use OS-specific icons for devices

Replace the generic bot avatar in device rows (and the chip) with the
machine's actual OS icon — Apple for darwin, Linux for linux, Microsoft
for win32, generic monitor as fallback. Matches the same icon set
already used in MCP plugin deployment.

* 💄 style(hetero-agent): unify execution targets into a single list

- Flatten This device / Cloud sandbox / remote devices into one list
- Add an info ⓘ icon in the popover header explaining when to pick a
  remote device vs This device; drop the inline section description
- Remove the "Other devices" rename and keep the original "Remote
  devices" terminology in the empty hint

* 💄 style(hetero-agent): rename popover title to Execution Device

* 💄 style(agent-signal): refine skill receipt card with self-evolution copy

- Render SkillsIcon for skill receipts and let PortalResourceCard accept a ReactNode icon
- Square 64x64 avatar, 12px corner radius, larger icon, drop the RadioTower marker
- Move the receipt card below the Usage row so it reads as metadata, not body content
- Reword the skill receipt to convey self-evolution ("Auto-learned a new skill" / "已自动习得新技能")

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

* 🐛 fix(hetero-agent): keep working-directory controls in RuntimeConfig

Revert the early-return I added in `RuntimeConfig.rightContent` for
heterogeneous agents. Hetero agents are mounted via `HeterogeneousChatInput`
which already replaces `RuntimeConfig` with `WorkingDirectoryBar` (where
the `HeteroDeviceSwitcher` lives), so the branch here was dead code — but
it dropped the `!isDesktop` gate and would have skipped the desktop
working-directory picker for any edge case that still falls through this
path (popup/share/embed). Restore the original web-only condition.

* ✨ feat(hetero-agent): fork dispatch by executionTarget for local CLI hetero

Local CLI hetero (claude-code / codex) used to dispatch to a device only
when topic.metadata.boundDeviceId was set, otherwise always spawned a
cloud sandbox — ignoring agencyConfig.executionTarget entirely.

Now resolve in this order:
  1. requestedDeviceId (topic-level override) → device dispatch, always wins
  2. agencyConfig.executionTarget = 'device' → dispatch to boundDeviceId;
     error out if no device is bound (no silent sandbox fallback, since
     the user explicitly chose this mode)
  3. otherwise (sandbox / local / unset) → cloud sandbox

'local' mode falls back to sandbox on the server since in-process spawn
only makes sense inside the Electron client; that path is owned by the
desktop and doesn't reach this code today.

* ✨ feat(hetero-agent): route runtime by executionTarget for local CLI hetero

Frontend complement to the previous server-side dispatch fork. Without
this change the chip's choice on desktop was a no-op: selectRuntimeType
hard-routed local CLI hetero to 'hetero' (desktop IPC) whenever
isDesktop, bypassing the server entirely — so 'device' / 'sandbox' picks
never reached the new server-side fork.

Now selectRuntimeType reads agencyConfig.executionTarget:
  - 'device'  → 'gateway' (server dispatches to bound lh connect device)
  - 'sandbox' → 'gateway' (server spawns cloud sandbox)
  - 'local'   → 'hetero' on desktop, 'gateway' on web (fallback)
  - unset     → legacy default (desktop = hetero, web = gateway)

All four runtime-selection call sites pass executionTarget through; the
non-hetero sub-agent dispatcher is unaffected since heteroProvider is
always undefined there.

* ✨ feat(chat-input): add Advanced Parameters entry to Plus menu

- New menu item toggles the right working sidebar's params tab, mirroring the agent header's ParamsPanelToggle
- Simplify the format-toolbar item label to a fixed "Show formatting toolbar" with a checkmark indicating active state
- Widen the active-label gap so the checkmark sits comfortably away from the text

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

* 🚩 feat(hetero-agent): gate execution-device switcher behind a lab flag

Add `enableExecutionDeviceSwitcher` to UserLabSchema (default off) and gate the heterogeneous WorkingDirectoryBar's HeteroDeviceSwitcher on it, so the new switcher can ship to canary without exposing it to all users until ready. Expose the toggle in Settings → Advanced → Labs.

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

---------

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