Skip to content

🐛 fix(agent-runtime): tighten isCanUseVision default and add aggregator fallback#14172

Merged
arvinxx merged 1 commit into
canaryfrom
arvinxx/fix/vision-capability-strict-default
Apr 25, 2026
Merged

🐛 fix(agent-runtime): tighten isCanUseVision default and add aggregator fallback#14172
arvinxx merged 1 commit into
canaryfrom
arvinxx/fix/vision-capability-strict-default

Conversation

@arvinxx

@arvinxx arvinxx commented Apr 25, 2026

Copy link
Copy Markdown
Member

💻 Change Type

  • 🐛 fix
  • ✅ test

🔗 Related Issue

Follow-up to LOBE-7214 — the original downgrade processor was correct, but the runtime capability probe was too permissive and prevented it from triggering for two real production cases.

🔀 Description of Change

createRuntimeExecutors in RuntimeExecutors.ts builds isCanUseVision by looking up the model card and returning info?.abilities?.vision ?? true. The ?? true fallback silently treated any model whose card omits the vision ability key as vision-capable, which neutralised the LOBE-7214 image-history downgrade for two clusters observed on the agent-gateway error dashboard:

  1. Models present in the registry without an explicit vision: true (e.g. the new deepseek-v4-pro card lists functionCall / reasoning / structuredOutput but no vision). info.abilities.vision is undefined, falls through to ?? true, and the harness happily forwards image_url content blocks to a text-only DeepSeek endpoint.
  2. Models routed via aggregator providers like lobehub, where the lookup (id, providerId) finds nothing because the registry only has the model under its upstream provider id (e.g. (deepseek-v4-pro, deepseek)). Same fallthrough.

Both surfaces produce the same upstream error:

ProviderBizError — Failed to deserialize the JSON body into the target type:
messages[N]: unknown variant `image_url`, expected `text`

This PR:

  • Switches the default to false, matching the existing isCanUseVideo convention. Models must explicitly opt in via abilities.vision = true to be treated as vision-capable.
  • Adds a cross-provider fallback so aggregator-routed model ids still resolve to their upstream model card (e.g. (claude-haiku-4-5, lobehub) falls back to the (claude-haiku-4-5, anthropic) entry).

🧪 How to Test

  • Added/updated tests

src/server/modules/AgentRuntime/__tests__/RuntimeExecutors.test.ts:

  • Updated (unknown, unknown) expectation from truefalse to match new default.
  • Added two cases for aggregator fallback: ('gpt-4', 'lobehub') resolves to true via the openai card, ('no-tools-model', 'lobehub') resolves to false.

📝 Additional Information

Production trace evidence (8 errors on 2026-04-25, all deepseek-v4-pro):

  • 2 via provider=lobehub — fall-through to default-true (case 2)
  • 6 via provider=deepseek — registry hit, but ability key missing (case 1)

After this fix:

  • (deepseek-v4-pro, deepseek) → registry hit, vision undefined, returns false
  • (deepseek-v4-pro, lobehub) → no direct hit, fallback to deepseek card, returns false
  • (claude-haiku-4-5, lobehub) → no direct hit, fallback to anthropic card with vision: true, returns true ✅ (no regression for vision models routed via aggregators)

… aggregator fallback

The runtime capability probe in RuntimeExecutors used `info?.abilities?.vision ?? true`,
which silently treated any model whose card omits the `vision` ability key as vision-capable.
This neutralised the LOBE-7214 downgrade pass for two real cases:

- Models present in the registry without an explicit `vision: true` (e.g. deepseek-v4-pro)
- Models routed through aggregator providers like `lobehub`, where `(model, providerId)` has
  no direct registry hit so the lookup fell through to the default

Switch the default to `false` (matching `isCanUseVideo`) and add a cross-provider fallback
that resolves an aggregator-routed model id against its upstream model card.

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

vercel Bot commented Apr 25, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
lobehub Ready Ready Preview, Comment Apr 25, 2026 0:57am

Request Review

@dosubot dosubot Bot added the size:S This PR changes 10-29 lines, ignoring generated files. label Apr 25, 2026

@sourcery-ai sourcery-ai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We've reviewed this pull request using the Sourcery rules engine

@codecov

codecov Bot commented Apr 25, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 67.59%. Comparing base (35c43fb) to head (69d61a4).
⚠️ Report is 3 commits behind head on canary.

Additional details and impacted files
@@            Coverage Diff            @@
##           canary   #14172     +/-   ##
=========================================
  Coverage   67.59%   67.59%             
=========================================
  Files        2192     2192             
  Lines      188999   188999             
  Branches    22503    18852   -3651     
=========================================
  Hits       127745   127745             
  Misses      61125    61125             
  Partials      129      129             
Flag Coverage Δ
app 60.62% <100.00%> (ø)
database 92.22% <ø> (ø)
packages/agent-runtime 79.82% <ø> (ø)
packages/context-engine 83.23% <ø> (ø)
packages/conversation-flow 92.40% <ø> (ø)
packages/file-loaders 87.02% <ø> (ø)
packages/memory-user-memory 74.74% <ø> (ø)
packages/model-bank 99.89% <ø> (ø)
packages/model-runtime 84.28% <ø> (ø)
packages/prompts 70.14% <ø> (ø)
packages/python-interpreter 92.90% <ø> (ø)
packages/ssrf-safe-fetch 0.00% <ø> (ø)
packages/utils 88.41% <ø> (ø)
packages/web-crawler 88.66% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Components Coverage Δ
Store 67.15% <ø> (ø)
Services 53.24% <ø> (ø)
Server 66.65% <100.00%> (ø)
Libs 53.30% <ø> (ø)
Utils 80.04% <ø> (ø)
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@arvinxx arvinxx merged commit 7c0203a into canary Apr 25, 2026
40 checks passed
@arvinxx arvinxx deleted the arvinxx/fix/vision-capability-strict-default branch April 25, 2026 13:22
Innei pushed a commit that referenced this pull request Apr 25, 2026
…or fallback (#14172)

🐛 fix(agent-runtime): tighten isCanUseVision default to false and add aggregator fallback

The runtime capability probe in RuntimeExecutors used `info?.abilities?.vision ?? true`,
which silently treated any model whose card omits the `vision` ability key as vision-capable.
This neutralised the LOBE-7214 downgrade pass for two real cases:

- Models present in the registry without an explicit `vision: true` (e.g. deepseek-v4-pro)
- Models routed through aggregator providers like `lobehub`, where `(model, providerId)` has
  no direct registry hit so the lookup fell through to the default

Switch the default to `false` (matching `isCanUseVideo`) and add a cross-provider fallback
that resolves an aggregator-routed model id against its upstream model card.

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@arvinxx arvinxx mentioned this pull request Apr 27, 2026
arvinxx added a commit that referenced this pull request Apr 27, 2026
# 🚀 LobeHub v2.1.53 (20260427)

**Release Date:** April 27, 2026
**Since v2.1.52:** 194 merged PRs · 17 contributors

> Introduce Heterogeneous Agent — Claude Code and Codex run as
first-class desktop runtimes, paired with a new Agent Signal package,
sharper desktop UX, and a wave of flagship model additions.

---

## ✨ Highlights

- **Introduce Heterogeneous Agent** — Claude Code and Codex run as
first-class desktop agents: subagent rendering, partial-message
streaming, multi-turn resume, terminal error surfacing, rich tool
inspectors, and runtime polish. (#14162, #13754, #14067, #14001, #13970,
#13942)
- **Screen capture & Quick Chat tray** — New desktop screen capture
overlay (macOS permission-gated) with Quick Chat tray and upload
pipeline improvements; chat input auto-focuses on overlay mount.
(#13818, #14097, #14105)
- **Desktop topic & tab UX** — Dedicated topic popup window with
cross-window sync, Cmd+W/Cmd+T tab shortcuts, TabBar polish, recent
working directories expanded to 20, and human approval notifications.
(#13957, #13983, #13972, #14036, #14092)
- **Git workflow built-in** — One-click pull/push from the branch chip,
ahead/behind badge, and submodule/worktree repo detection. (#14041,
#13980, #13978)
- **Agent Signal package** — New `@lobechat/agent-signal` runtime for
dynamic memory feedback signals, with OTel metrics and self-iteration in
Lab. (#14157, #14170, #14159, #14169, #14187)
- **New models** — Claude Opus 4.7 with `xhigh` effort tier, GPT-5.5,
DeepSeek V4 Flash/Pro with reasoning slider, Kimi K2.6, MiMo-V2.5/Pro,
gpt-image-2, Qwen3.6 Flash/Plus, and Pixverse-c1. (#13903, #14147,
#14114, #14004, #14089, #14039, #13923)
- **New providers** — OpenCode Zen, OpenCode Go, and Azure OpenAI Router
runtime. (#13943, #14064, #13823)
- **Mobile settings overhaul** — Full settings menu and responsive
profile layout for mobile. (#14019)

---

## 🏗️ Heterogeneous Agent

- Claude Code runtime, working-directory awareness, and sidebar polish.
(#13970)
- CC subagent rendering with persistent streamed text; parallel-tool
orphan fix. (#14001, #13968, #14024)
- Per-step usage persisted to each step assistant message. (#13964)
- Per-phase workflow expand defaults; full-expand toggle with
three-level expansion. (#14171, #13906)
- Hetero-mode actions bar; tool inspector polish. (#13963, #14034,
#14030)
- Codex desktop integration with rich tool rendering and devtools
preview. (#14067, #14100)
- Codex terminal error surfacing and CLI output tracing. (#14166)
- Tighten `isCanUseVision` default and add aggregator fallback. (#14172)
- Persist `ccSessionId` in topic metadata for CC multi-turn resume.
(#13902)
- CC account card, topic filter, and integration polish. (#13955,
#13942, #13950)
- Token-level deltas streamed via `--include-partial-messages`. (#13929)

---

## 🧠 Agent Signal & Self-Iteration

- New `@lobechat/agent-signal` package with dynamic feedback signals.
(#14157)
- AgentSignalRuntime wired through agent-tracing and observability-otel
metrics. (#14170, #14159)
- Self-iteration feature flag added to Lab; front-side flag check.
(#14169, #14186)
- Signal policy for receiving memory feedback dynamically. (#14187)

---

## 💬 Conversation

- Queue follow-up sends during running CC turns. (#14179)
- Persist per-topic chat scroll position; pin user message + fold long
messages. (#14191, #14056)
- Inline resend when editing last user message. (#14080)
- Disable first-block markdown streaming to prevent flicker. (#14193,
#13904)
- Prevent Markdown stream replay when vlist remounts streaming items.
(#14086)
- Stop repinning after manual scroll; unify scroll-to-user + spacer
hooks. (#14099, #14132)

---

## 📱 Platforms & Integrations

### Desktop / Electron

- Screen capture overlay, Quick Chat tray, and upload pipeline
improvements. (#13818)
- macOS permission gate for screen capture; auto-focus chat panel input.
(#14097, #14105)
- Dedicated topic popup window with cross-window sync. (#13957)
- TabBar polish: `+` button for new topic, dark theme blend, close icon
by default. (#13972, #14203, #13973)
- Recent working directories expanded from 5 to 20; submodule/worktree
repo detection. (#14036, #13978)
- Cmd+W / Cmd+T tab shortcuts and global shortcut consolidation.
(#13983, #13880)
- Linux icon configuration; human approval desktop notifications.
(#14042, #14092)

### Git Workflow

- One-click pull/push from branch chip; ahead/behind badge with
refactored GitCtr. (#14041, #13980)

### Mobile

- Full settings menu and responsive profile layout. (#14019)
- Agent route added to mobile router; mobile agent topic route
registered. (#14103, #14158)
- Session list skeleton row layout corrected. (#14040)

### Bot / Messaging

- DM strategy support; bot emoji and markdown render optimization.
(#14201, #14091, #14140)
- Slack webhook fix; bot platform setup guide reference. (#14052,
#14121)

---

## 🤖 Models & Providers

### New models

- **Claude Opus 4.7** with `xhigh` effort tier; strip temperature/top_p.
(#13903, #13909)
- **GPT-5.5**. (#14147)
- **DeepSeek V4** Flash/Pro cards with reasoning slider; cache-hit and
Pro discount pricing. (#14114, #14209, #14196, #14131)
- **Kimi K2.6** model with LobeHub-hosted card. (#14004, #14006)
- **MiMo-V2.5 / V2.5-Pro**. (#14089)
- **gpt-image-2**, **Qwen3.6 Flash/Plus**, **Pixverse-c1**. (#14039,
#13923)

### New providers

- **OpenCode Zen** and **OpenCode Go** with env-var support. (#13943,
#14064)
- **Azure OpenAI Router** runtime support. (#13823)
- Model alias mapping for image and video runtimes. (#13896)
- Seedance video models migrated to Dreamina. (#14144)

### Runtime reliability

- Sanitize invalid tool_call arguments to unbreak strict providers.
(#14033)
- Tolerate null `function.name` in streaming tool_call deltas. (#14139)
- Preserve Gemini 3 `thoughtSignature` in `call_tools_batch`
normalization. (#14032)
- Downgrade `image_url` parts when target model lacks vision. (#14029)
- Preserve Cloudflare provider error context. (#14136)
- Use `safety_identifier` for OpenAI Responses API. (#14148)
- Unwrap underlying PG error in `formatErrorEventData`. (#14038)

---

## 🖥️ User Experience

- **Onboarding** — Preset agent naming suggestions, structured hunk ops
for `updateDocument`, persona analytics snapshot, footer promotion
pipeline, wrap-up button. (#13931, #13989, #13930, #13853, #13934)
- **Document workflow** — Agent documents promoted as primary workspace
panel; history management and compare workflow; web-crawl docs
associated with agent documents. (#13924, #13725, #13893)
- **cmdk** — Agent identity surfaced on topic search results;
topic/message search scoped to current agent. (#14204, #13960)
- **Floating chat panel** and workspace improvements. (#13887)
- **Topic completion status** with dropdown action and filter. (#14005)

---

## 🔧 Tooling

- Redis-backed feature flag provider for runtime config. (#14098)
- Vite upgraded to 8.0.0 with Rolldown strict execution order. (#12720,
#14058)
- `@lobechat/model-bank` automated npm release with provenance. (#14015,
#14017, #14018)
- Skill activation fallback when `activateTools` cannot find identifier.
(#14010)
- Cron tool: timezone and existing jobs injected into system prompt;
clarified `lobe-gtd` and `lobe-cron` descriptions. (#14012, #14013)

---

## 🔒 Security & Reliability

- **Security:** uuid bumped to v14 (advisory). (#14083)
- **Security:** validate avatar URL and scope old-avatar deletion to
owner. (#13982)
- **Security:** clear OIDC sessions on better-auth signout; return 401
(not 500) for expired OIDC JWT. (#13916, #14014)
- **Reliability:** scope pending-approval check to current assistant
turn. (#14182)
- **Reliability:** sanitize heterogeneous-agent attachment cache
filenames. (#13937)
- **Reliability:** reduce subagent task status error noise. (#14026)

---

## 👥 Contributors

Huge thanks to **17 contributors** who shipped **194 merged PRs** this
week.

@hardy · @shaun0927 · @hezhijie0327 · @sxjeru · @arvinxx · @Innei ·
@tjx666 · @lijian · @neko · @rdmclin2 · @AmAzing129 · @sudongyuer ·
@CanisMinor · @rivertwilight

Plus @lobehubbot and renovate[bot] for maintenance.

---

**Full Changelog**:
v2.1.52...v2.1.53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

size:S This PR changes 10-29 lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant