repo stringclasses 1
value | github_id int64 3.77B 4.38B | github_node_id stringlengths 18 24 | number int64 75 77.8k | html_url stringlengths 46 49 | api_url stringlengths 56 59 | title stringlengths 1 746 | body stringlengths 1 55.7k ⌀ | state stringclasses 2
values | state_reason stringclasses 4
values | locked bool 2
classes | comments_count int64 0 128 | labels listlengths 0 5 | assignees listlengths 0 2 | created_at stringdate 2026-01-01 16:55:34 2026-05-05 11:59:03 | updated_at stringdate 2026-03-24 00:13:29 2026-05-05 11:59:15 | closed_at stringlengths 20 20 ⌀ | author_association stringclasses 3
values | milestone_title stringclasses 0
values | snapshot_id stringclasses 76
values | extracted_at stringdate 2026-03-24 23:36:49 2026-05-05 12:00:33 | author_login stringlengths 3 38 | author_id int64 346 282M | author_node_id stringlengths 12 20 | author_type stringclasses 2
values | author_site_admin bool 2
classes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
openclaw/openclaw | 4,368,500,963 | I_kwDOQb6kR88AAAABBGII4w | 76,013 | https://github.com/openclaw/openclaw/issues/76013 | https://api.github.com/repos/openclaw/openclaw/issues/76013 | [Bug]: Telegram channel: fetch-timeout 10000ms on sendChatAction/sendMessage (undici, Windows, Node 24) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
On Windows with Node.js 24, the OpenClaw Telegram channel fails to deliver
agent replies because undici HTTP requests to api.telegram.org exceed the
hardcoded 10000ms fetch timeout, even though the Telegram API itself resp... | closed | completed | true | 2 | [
"bug",
"regression"
] | [] | 2026-05-02T08:57:28Z | 2026-05-05T04:27:49Z | 2026-05-02T10:01:27Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | iaki1206 | 17,490,720 | MDQ6VXNlcjE3NDkwNzIw | User | false |
openclaw/openclaw | 4,368,424,894 | I_kwDOQb6kR88AAAABBGDfvg | 75,999 | https://github.com/openclaw/openclaw/issues/75999 | https://api.github.com/repos/openclaw/openclaw/issues/75999 | [Bug]: 4.29 dispatch prep stages take ~73s of synchronous CPU work, blocking event loop | # [Bug]: 4.29 dispatch prep stages take ~73s of synchronous CPU work, blocking event loop
## Bug type
Performance regression (introduced in 4.29; not present in 4.27)
## Summary
Upgrading from 4.24/4.27 → 4.29 caused every agent dispatch to take **2–5 minutes** to first reply. The gateway log shows new `prep stages... | closed | completed | true | 12 | [] | [] | 2026-05-02T08:16:22Z | 2026-05-05T04:26:17Z | 2026-05-02T12:48:22Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | zackchiutw | 107,488,700 | U_kgDOBmglvA | User | false |
openclaw/openclaw | 4,368,528,088 | I_kwDOQb6kR88AAAABBGJy2A | 76,015 | https://github.com/openclaw/openclaw/issues/76015 | https://api.github.com/repos/openclaw/openclaw/issues/76015 | [Bug]: Performance regression in v2026.4.29: embedded-run prep stalls and gateway event-loop blockage | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from v2026.4.27 to v2026.4.29 in a Docker-based NAS deployment, gateway responsiveness became much worse, with embedded-run startup/prep delays and liveness warnings showing event-loop blockage.
### Steps t... | closed | not_planned | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T09:12:32Z | 2026-05-05T04:27:56Z | 2026-05-02T09:17:24Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | realistic-guy | 59,161,385 | MDQ6VXNlcjU5MTYxMzg1 | User | false |
openclaw/openclaw | 4,361,552,521 | I_kwDOQb6kR88AAAABA_gCiQ | 75,283 | https://github.com/openclaw/openclaw/issues/75283 | https://api.github.com/repos/openclaw/openclaw/issues/75283 | 2026.4.29: Gateway repeatedly spawns runtime-deps pnpm installs and starves event loop | Supersedes #75275. This is the clean issue with the later runtime-deps evidence included directly in the main body.
### Bug type
Performance / hang / runtime-deps reinstall loop / gateway health regression
### Summary
After upgrading the same Ubuntu VPS that reproduced #75259 to the stable `2026.4.29` package, the ... | closed | completed | true | 9 | [] | [] | 2026-04-30T22:07:29Z | 2026-05-04T04:33:41Z | 2026-05-01T07:37:57Z | CONTRIBUTOR | null | 20260504T060031Z | 2026-05-04T06:00:31Z | brokemac79 | 255,583,030 | U_kgDODzvjNg | User | false |
openclaw/openclaw | 4,368,573,496 | I_kwDOQb6kR88AAAABBGMkOA | 76,023 | https://github.com/openclaw/openclaw/issues/76023 | https://api.github.com/repos/openclaw/openclaw/issues/76023 | Regression: embedded agent prep takes 50–80s on 2026.4.29 even with only one channel plugin enabled | ### Summary
After upgrading to `2026.4.29`, embedded agent startup/prep became extremely slow. Messages are received by the channel promptly, but the embedded agent does not become `stream-ready` for roughly **50–80 seconds**.
This appears to be in the gateway / embedded agent prep path rather than the Weixin channel... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T09:38:37Z | 2026-05-05T04:27:52Z | 2026-05-02T09:42:43Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | CharyeahOwO | 105,187,600 | U_kgDOBkUJEA | User | false |
openclaw/openclaw | 4,368,604,809 | I_kwDOQb6kR88AAAABBGOeiQ | 76,030 | https://github.com/openclaw/openclaw/issues/76030 | https://api.github.com/repos/openclaw/openclaw/issues/76030 | [Bug]: Skill symlinks in extraDirs roots cause repeated symlink-escape errors and CPU overhead | ⚠️ Filed by OpenClaw Agent (echo) on behalf of stevenliou.
---
## Bug Report
**Filed by:** OpenClaw Agent (echo) on behalf of stevenliou
### Summary
Placing symlinks to skills inside a configured extraDirs root inside another extraDirs root causes thousands of errors on every gateway startup, creating measurable ... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T09:56:18Z | 2026-05-05T04:27:51Z | 2026-05-02T09:59:34Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | stevenliou | 9,360,792 | MDQ6VXNlcjkzNjA3OTI= | User | false |
openclaw/openclaw | 4,368,601,711 | I_kwDOQb6kR88AAAABBGOSbw | 76,029 | https://github.com/openclaw/openclaw/issues/76029 | https://api.github.com/repos/openclaw/openclaw/issues/76029 | [Feature]: [4.29] Gateway startup hangs due to blocking provider auth (codex token exchange with no timeout) | ### Summary
Provider auth during startup blocks event loop with no timeout when endpoint is unreachable
### Problem to solve
When the gateway starts, it probes every configured provider — even unused ones — and performs an auth/token exchange synchronously on the main thread.
If a provider uses “auth: token” (e.g.,... | closed | completed | true | 1 | [
"enhancement"
] | [] | 2026-05-02T09:54:29Z | 2026-05-05T04:27:51Z | 2026-05-02T09:59:47Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Wchy520 | 130,220,750 | U_kgDOB8MCzg | User | false |
openclaw/openclaw | 4,368,552,053 | I_kwDOQb6kR88AAAABBGLQdQ | 76,018 | https://github.com/openclaw/openclaw/issues/76018 | https://api.github.com/repos/openclaw/openclaw/issues/76018 | [Bug]: DeepSeek V4 via OpenRouter fails on follow-up/tool turns with reasoning_content error because DeepSeek wrapper only applies to provider=deepseek | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
When using `deepseek/deepseek-v4-flash` through OpenRouter in OpenClaw, follow-up turns can fail with:
`code: 20015`
`The reasoning_content in the thinking mode must be passed back to the API.`
This appears to happen becau... | closed | completed | true | 2 | [
"bug",
"regression"
] | [] | 2026-05-02T09:26:16Z | 2026-05-05T04:27:45Z | 2026-05-02T10:16:40Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | cloph-dsp | 128,580,843 | U_kgDOB6n86w | User | false |
openclaw/openclaw | 4,368,545,711 | I_kwDOQb6kR88AAAABBGK3rw | 76,017 | https://github.com/openclaw/openclaw/issues/76017 | https://api.github.com/repos/openclaw/openclaw/issues/76017 | doctor: Gateway PATH check reports false warnings for non-existent user directories | ## Problem
Running `openclaw doctor` shows this warning:
```
Gateway service PATH missing required dirs:
/Users/xiphis/.npm-global/bin, /Users/xiphis/bin,
/Users/xiphis/.nix-profile/bin
```
These directories **do not exist** on my system and **never have**. They are not in my shell PATH, not in the Gateway service P... | closed | completed | true | 2 | [] | [] | 2026-05-02T09:22:34Z | 2026-05-05T04:27:41Z | 2026-05-02T10:37:43Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | sysit | 9,957,218 | MDQ6VXNlcjk5NTcyMTg= | User | false |
openclaw/openclaw | 4,368,677,413 | I_kwDOQb6kR88AAAABBGS6JQ | 76,038 | https://github.com/openclaw/openclaw/issues/76038 | https://api.github.com/repos/openclaw/openclaw/issues/76038 | [Bug]: Stuck Session Recovery 机制双重失效 + Session 预处理每次耗时过长 | ## Bug 报告
### 🔴 核心 Bug：Stuck Session Recovery 机制双重失效
**问题描述**：
Session 长时间停留在 \`processing\` 状态，事件循环完全阻塞，导致 Gateway 无响应，最终被 systemd 超时强杀。
**两种情况均不恢复（根因）**：
情况 A — 有活跃工作时：
\`\`\`
stuck session recovery skipped: reason=active_reply_work action=keep_lane
age=289s, 385s... 持续多分钟，系统选择"保持 lane 占用"不处理
\`\`\`
情况 B — 无活跃工... | open | null | false | 3 | [] | [] | 2026-05-02T10:36:29Z | 2026-05-04T16:58:05Z | null | NONE | null | 20260504T180039Z | 2026-05-04T18:00:39Z | WS-Q0758 | 274,219,387 | U_kgDOEFhBew | User | false |
openclaw/openclaw | 4,368,641,515 | I_kwDOQb6kR88AAAABBGQt6w | 76,035 | https://github.com/openclaw/openclaw/issues/76035 | https://api.github.com/repos/openclaw/openclaw/issues/76035 | Slack message hooks lose bot alert text when command body is blank |
### Problem
Slack bot messages can have empty top-level `text` while carrying the user-visible message in attachment text. OpenClaw's Slack prepare path extracts that text into `RawBody` and `BodyForAgent`, so the agent can reply. The plugin hook mapper previously treated an empty string `BodyForCommands` as the cano... | closed | completed | true | 1 | [] | [] | 2026-05-02T10:16:08Z | 2026-05-05T04:27:39Z | 2026-05-02T10:43:48Z | CONTRIBUTOR | null | 20260505T060034Z | 2026-05-05T06:00:34Z | amsminn | 66,085,474 | MDQ6VXNlcjY2MDg1NDc0 | User | false |
openclaw/openclaw | 4,368,691,720 | I_kwDOQb6kR88AAAABBGTyCA | 76,041 | https://github.com/openclaw/openclaw/issues/76041 | https://api.github.com/repos/openclaw/openclaw/issues/76041 | [Bug] OpenClaw 2026.4.29 Windows: 162s eventLoopDelay blocking Gateway | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Gateway eventLoop completely blocked on every message. Single "状态如何" query takes 8 minutes to respond. eventLoopDelayP99Ms=162403.5ms, core-plugin-tools=162299ms, system-prompt=81039ms, stream-setup=77501ms. MiniMax API dire... | closed | completed | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T10:43:45Z | 2026-05-05T04:27:38Z | 2026-05-02T10:48:36Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | gaozhaohe050-hue | 281,147,217 | U_kgDOEMH3UQ | User | false |
openclaw/openclaw | 4,368,703,655 | I_kwDOQb6kR88AAAABBGUgpw | 76,042 | https://github.com/openclaw/openclaw/issues/76042 | https://api.github.com/repos/openclaw/openclaw/issues/76042 | [Bug]: Upgrade or Clean install is VERY long or broken. Right now, ater 2 hours is not finished. | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Previous versions 2026.3.xx i can install in ~20 min. Onboarding 2-3 min. Tested on different laptops, weak harware etc.
Right now i am not able to start 2026.4.29 in normal waiting time, i can wait 5 min, but not more.
Exa... | open | null | false | 4 | [
"bug",
"regression"
] | [
"vincentkoc"
] | 2026-05-02T10:50:48Z | 2026-05-03T10:20:22Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | danilovmy | 39,524,090 | MDQ6VXNlcjM5NTI0MDkw | User | false |
openclaw/openclaw | 4,368,712,979 | I_kwDOQb6kR88AAAABBGVFEw | 76,043 | https://github.com/openclaw/openclaw/issues/76043 | https://api.github.com/repos/openclaw/openclaw/issues/76043 | Active-memory embedded run startup overhead: 44-51s before LLM call, 100% timeout rate | # OpenClaw Active-Memory Bug Report: Embedded Run Startup Overhead
## Summary
Active-memory plugin has **0% success rate** due to consistent 45-second timeouts. Investigation reveals the bottleneck is NOT the model or memory search, but OpenClaw's embedded run framework's startup overhead.
## Impact
- Active-memory... | closed | duplicate | true | 2 | [] | [] | 2026-05-02T10:56:24Z | 2026-05-05T04:27:22Z | 2026-05-02T12:10:55Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | benputley1 | 110,424,063 | U_kgDOBpTv_w | User | false |
openclaw/openclaw | 4,368,731,732 | I_kwDOQb6kR88AAAABBGWOVA | 76,045 | https://github.com/openclaw/openclaw/issues/76045 | https://api.github.com/repos/openclaw/openclaw/issues/76045 | Feishu: message tool presentation does not render as interactive card | ## Bug Description
When using the `message` tool with the `presentation` parameter on Feishu channel, the output degrades to plain text instead of rendering as an interactive card with buttons.
## Expected Behavior
`message` tool with `presentation` containing buttons should render as a Feishu interactive card with ... | open | null | false | 1 | [] | [] | 2026-05-02T11:05:54Z | 2026-05-03T09:13:46Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | fanxlab | 231,185,438 | U_kgDODcecHg | User | false |
openclaw/openclaw | 4,368,571,093 | I_kwDOQb6kR88AAAABBGMa1Q | 76,022 | https://github.com/openclaw/openclaw/issues/76022 | https://api.github.com/repos/openclaw/openclaw/issues/76022 | [Bug]: 2026.04.29 installation：failed to load bundled channels 安装报错：信道配置全部丢失 | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from 4.27 to 4.29 by 'openclaw update', it takes about 5 mins for the automatic doctor post installation diagnosis, and 'openclaw status' shows the follwoing message after the 1st post-installation run;
执行 o... | closed | not_planned | true | 1 | [
"bug"
] | [] | 2026-05-02T09:37:08Z | 2026-05-05T04:27:29Z | 2026-05-02T11:24:43Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | yaaboo-gif | 254,105,470 | U_kgDODyVXfg | User | false |
openclaw/openclaw | 4,368,760,462 | I_kwDOQb6kR88AAAABBGX-jg | 76,047 | https://github.com/openclaw/openclaw/issues/76047 | https://api.github.com/repos/openclaw/openclaw/issues/76047 | Two issues causing persistent event loop saturation: temp file leak + node.list probe timeout | ## Summary
Two related issues that together cause `eventLoopUtilization=1` and `cpuCoreRatio=1.5+`, making the gateway unresponsive (cron tool timeouts, slow responses).
## Issue 1: Temp files not cleaned up after atomic writes
`.tmp` files accumulate indefinitely after atomic writes:
- `~/.openclaw/devices/paired.j... | open | null | false | 1 | [] | [] | 2026-05-02T11:21:50Z | 2026-05-03T08:26:57Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | stephan-afk | 281,158,632 | U_kgDOEMIj6A | User | false |
openclaw/openclaw | 4,368,768,078 | I_kwDOQb6kR88AAAABBGYcTg | 76,050 | https://github.com/openclaw/openclaw/issues/76050 | https://api.github.com/repos/openclaw/openclaw/issues/76050 | Bug: Control UI sessions list defaults can trigger expensive full-history scans | ## Summary
The Control UI Sessions tab currently initializes with an unbounded `activeMinutes` filter and a default limit of `120`. On instances with a larger session history, this can make the first `sessions.list` call scan too much history and periodically block the gateway/UI for several seconds.
A safer default ... | closed | completed | true | 2 | [] | [] | 2026-05-02T11:26:16Z | 2026-05-05T04:26:57Z | 2026-05-02T14:20:26Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Neomail2 | 120,675,713 | U_kgDOBzFdgQ | User | false |
openclaw/openclaw | 4,368,683,693 | I_kwDOQb6kR88AAAABBGTSrQ | 76,040 | https://github.com/openclaw/openclaw/issues/76040 | https://api.github.com/repos/openclaw/openclaw/issues/76040 | [Feature]: My request | ### Summary
I have completed your test project.
### Problem to solve
I received your test project on April 21st.
### Proposed solution
So, I finished it on April 22nd, sent it, and sent an invitation to my project.
### Alternatives considered
_No response_
### Impact
However, since I haven't heard anything yet... | closed | not_planned | true | 1 | [
"enhancement"
] | [] | 2026-05-02T10:39:08Z | 2026-05-05T04:26:49Z | 2026-05-02T10:50:11Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | farooquisaud | 42,851,100 | MDQ6VXNlcjQyODUxMTAw | User | false |
openclaw/openclaw | 4,368,677,838 | I_kwDOQb6kR88AAAABBGS7zg | 76,039 | https://github.com/openclaw/openclaw/issues/76039 | https://api.github.com/repos/openclaw/openclaw/issues/76039 | [Feature]: My request | ### Summary
I have completed your test project.
### Problem to solve
I received your test project on April 21st.
### Proposed solution
So, I finished it on April 22nd, sent it, and sent an invitation to my project.
### Alternatives considered
_No response_
### Impact
However, since I haven't heard anything yet... | closed | not_planned | true | 1 | [
"enhancement"
] | [] | 2026-05-02T10:36:41Z | 2026-05-05T04:27:28Z | 2026-05-02T10:50:07Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | farooquisaud | 42,851,100 | MDQ6VXNlcjQyODUxMTAw | User | false |
openclaw/openclaw | 4,368,740,991 | I_kwDOQb6kR88AAAABBGWyfw | 76,046 | https://github.com/openclaw/openclaw/issues/76046 | https://api.github.com/repos/openclaw/openclaw/issues/76046 | Per-agent contextInjection override | ## Problem
`agents.defaults.contextInjection` is the only place this setting can be configured. There is no per-agent override (`agents.list[].contextInjection`).
This becomes load-bearing for setups that run multiple agents with different rule-strictness needs:
- A general-purpose `main` agent benefits from `contin... | open | null | false | 1 | [] | [] | 2026-05-02T11:10:47Z | 2026-05-03T06:51:05Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | lucidlearning314 | 277,298,068 | U_kgDOEIc7lA | User | false |
openclaw/openclaw | 4,368,765,322 | I_kwDOQb6kR88AAAABBGYRig | 76,048 | https://github.com/openclaw/openclaw/issues/76048 | https://api.github.com/repos/openclaw/openclaw/issues/76048 | [Bug]: ZAI GLM-5 reasoning models return empty responses — reasoning_content routed to thinking instead of text (regression from 2026.4.24+) | ## Summary
ZAI GLM-5 reasoning models (`glm-5-turbo`, `glm-5.1`) return empty/incoherent responses in OpenClaw 2026.4.29. The API returns all actual text in `reasoning_content` field with `content=""`, but the stream processing pipeline routes `reasoning_content` to `appendThinkingDelta()` (hidden) instead of `appendT... | open | null | false | 3 | [] | [] | 2026-05-02T11:24:38Z | 2026-05-03T11:50:53Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | Timluogit | 49,114,690 | MDQ6VXNlcjQ5MTE0Njkw | User | false |
openclaw/openclaw | 4,368,800,792 | I_kwDOQb6kR88AAAABBGacGA | 76,059 | https://github.com/openclaw/openclaw/issues/76059 | https://api.github.com/repos/openclaw/openclaw/issues/76059 | [Bug]: Web UI msg-meta contextPercent shows 100% after long sessions due to _j() accumulating cacheRead across messages |
## Bug Description
After a long session (821 assistant messages, ~20 hours of conversation), the **Web UI message metadata bar** shows **100% ctx** with a red/danger indicator. This makes it look like the 1M context window is full, but the actual context usage is only ~12.8%.
## Root Cause
In `control-ui/assets/ind... | closed | completed | true | 1 | [] | [] | 2026-05-02T11:42:23Z | 2026-05-05T04:27:25Z | 2026-05-02T12:03:36Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | zdwalter | 360,915 | MDQ6VXNlcjM2MDkxNQ== | User | false |
openclaw/openclaw | 4,368,796,067 | I_kwDOQb6kR88AAAABBGaJow | 76,057 | https://github.com/openclaw/openclaw/issues/76057 | https://api.github.com/repos/openclaw/openclaw/issues/76057 | [Bug]: Gateway CPU 60-80% on ARM64 Linux, event loop blocked 37s (v2026.4.29) | ## [Bug]: Gateway CPU 60-80% on ARM64 Linux, event loop blocked 37s (v2026.4.29)
### Bug type
Behavior bug (incorrect output/state without crash)
### Summary
Gateway process consumes 60-80% CPU on ARM64 Linux (aarch64) with v2026.4.29, even with minimal activity. Event loop is blocked up to 37-40 seconds per agent ... | closed | completed | true | 5 | [] | [] | 2026-05-02T11:39:56Z | 2026-05-05T04:27:13Z | 2026-05-02T13:28:20Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | zjx19 | 172,806,122 | U_kgDOCkzP6g | User | false |
openclaw/openclaw | 4,368,807,955 | I_kwDOQb6kR88AAAABBGa4Ew | 76,061 | https://github.com/openclaw/openclaw/issues/76061 | https://api.github.com/repos/openclaw/openclaw/issues/76061 | Per-request startup pipeline takes 90-130s (core-plugin-tools + system-prompt + stream-setup) | ## Environment
- OpenClaw: v2026.4.29
- Node.js: v24.14.0
- Host: Ubuntu 22.04, 16GB RAM, 8 vCPU (no GPU)
- Plugins: feishu, memory-core
- Workspace: 6 skills, ~12KB md files
- Primary model: volcengine-plan/glm-5.1 (remote)
## Problem
Every user message requires 90-130 seconds before the first token streams. trace:em... | closed | not_planned | true | 2 | [] | [] | 2026-05-02T11:45:41Z | 2026-05-05T04:27:20Z | 2026-05-02T12:19:22Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | STLI69 | 221,152,097 | U_kgDODS6DYQ | User | false |
openclaw/openclaw | 4,368,809,267 | I_kwDOQb6kR88AAAABBGa9Mw | 76,062 | https://github.com/openclaw/openclaw/issues/76062 | https://api.github.com/repos/openclaw/openclaw/issues/76062 | [Bug]: QQ Bot C2C images downloaded successfully but attachments are missing from inbound message to Agent | # [Bug]: QQ Bot C2C images downloaded successfully but attachments are missing from inbound message to Agent
## Summary
When a user sends an image via QQ Bot C2C (private chat), the image file is downloaded to `~/.openclaw/media/qqbot/downloads/` successfully, but the Agent does **not** receive the attachment metadat... | open | null | false | 1 | [] | [] | 2026-05-02T11:46:16Z | 2026-05-03T06:51:06Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | bxy3045134656 | 183,266,172 | U_kgDOCuxrfA | User | false |
openclaw/openclaw | 4,368,815,180 | I_kwDOQb6kR88AAAABBGbUTA | 76,063 | https://github.com/openclaw/openclaw/issues/76063 | https://api.github.com/repos/openclaw/openclaw/issues/76063 | MCP server tools missing from agent request body — regression still present in v2026.4.27 | ## Summary
MCP server tools are not included in the agent's outbound request body in v2026.4.27. The regression was marked as fixed in v2026.4.27 but testing confirms the issue persists. v2026.4.27 also introduces silent confabulation as the failure mode, more dangerous than the honest failure in v2026.4.26.
## Envir... | open | null | false | 1 | [] | [] | 2026-05-02T11:49:57Z | 2026-05-03T06:51:08Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | luuq2026 | 268,951,258 | U_kgDOEAfe2g | User | false |
openclaw/openclaw | 4,368,931,138 | I_kwDOQb6kR88AAAABBGiZQg | 76,074 | https://github.com/openclaw/openclaw/issues/76074 | https://api.github.com/repos/openclaw/openclaw/issues/76074 | docs(capabilities): define workspace capability discovery convention | ## Summary
Define a workspace capability discovery convention so agents can discover local, durable capabilities before concluding that a task is impossible.
## Problem to solve
OpenClaw already has several primitives that can support this behavior, including bootstrap hooks, `before_prompt_build`, taskflow, ACP run... | open | null | false | 1 | [] | [] | 2026-05-02T12:50:11Z | 2026-05-03T11:26:11Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | MertBasar0 | 100,199,793 | U_kgDOBfjtcQ | User | false |
openclaw/openclaw | 4,369,025,651 | I_kwDOQb6kR88AAAABBGoKcw | 76,087 | https://github.com/openclaw/openclaw/issues/76087 | https://api.github.com/repos/openclaw/openclaw/issues/76087 | [Bug] Restart-sentinel continuation can get stuck after transient Telegram sendMessage failure | ## Summary
Restart-sentinel continuation delivery can become stuck in `~/.openclaw/session-delivery-queue/` after a transient Telegram `sendMessage` network failure. The continuation entry is persisted and retry metadata is updated, but there appears to be no periodic/session wake recovery after the immediate startup ... | open | null | false | 1 | [] | [] | 2026-05-02T13:33:43Z | 2026-05-03T10:20:24Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | richardmqq | 14,178,267 | MDQ6VXNlcjE0MTc4MjY3 | User | false |
openclaw/openclaw | 4,368,976,935 | I_kwDOQb6kR88AAAABBGlMJw | 76,076 | https://github.com/openclaw/openclaw/issues/76076 | https://api.github.com/repos/openclaw/openclaw/issues/76076 | [Feature]: wechat-access 插件目前不支持微信群聊 | ### Summary
## 功能建议：wechat-access 插件支持微信群聊
### 现状
我在使用 QClaw 的 wechat-access 插件配置微信个人通道，体验很顺畅。
目前遇到一个问题：希望能够在**微信群**里实现 AI 自动回复功能。
### Problem to solve
在微信群里作为 AI 助手，自动回答群友问题
配合定时任务做群消息监控和提醒
### Proposed solution
请问
是否有支持微信群聊的计划？
是否有变通的方案可以先实现群聊功能？
### Alternatives considered
_No response_
### Impact
微信群是国内最... | closed | not_planned | true | 1 | [
"enhancement"
] | [] | 2026-05-02T13:10:09Z | 2026-05-05T04:27:12Z | 2026-05-02T13:38:53Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | BluewhALE087 | 272,141,397 | U_kgDOEDiMVQ | User | false |
openclaw/openclaw | 4,369,031,468 | I_kwDOQb6kR88AAAABBGohLA | 76,089 | https://github.com/openclaw/openclaw/issues/76089 | https://api.github.com/repos/openclaw/openclaw/issues/76089 | [Feature]: I want a Spawn a sub-agent button in web dashboard ui | ### Summary
I want a Spawn a sub-agent button in web dashboard ui
### Problem to solve
I want a Spawn a sub-agent button in web dashboard ui
### Proposed solution
I want a Spawn a sub-agent button in web dashboard ui
### Alternatives considered
_No response_
### Impact
I want a Spawn a sub-agent button in web ... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-05-02T13:36:01Z | 2026-05-03T11:50:55Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | cyrus123456 | 73,327,722 | MDQ6VXNlcjczMzI3NzIy | User | false |
openclaw/openclaw | 4,369,027,806 | I_kwDOQb6kR88AAAABBGoS3g | 76,088 | https://github.com/openclaw/openclaw/issues/76088 | https://api.github.com/repos/openclaw/openclaw/issues/76088 | After upgrade to 4.29: Gateway starts but all bundled channels crash — missing json5 in plugin-runtime-deps | ## Bug Description
After upgrading from 2026.4.26 to 2026.4.29 (and also reproducible on 4.26), all bundled channel plugins (Telegram, Discord, WhatsApp) crash on startup with:
```
Cannot find package 'json5' imported from
~/.openclaw/plugin-runtime-deps/openclaw-2026.4.29-96928547ff78/dist/frontmatter-Cc-V8aI2.js... | closed | completed | true | 2 | [] | [] | 2026-05-02T13:34:34Z | 2026-05-05T04:27:11Z | 2026-05-02T13:39:13Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | gzxmren | 41,671,592 | MDQ6VXNlcjQxNjcxNTky | User | false |
openclaw/openclaw | 4,369,072,769 | I_kwDOQb6kR88AAAABBGrCgQ | 76,095 | https://github.com/openclaw/openclaw/issues/76095 | https://api.github.com/repos/openclaw/openclaw/issues/76095 | [Bug]: openclaw-weixin channel plugin missing channelConfigs metadata causes duplicate startup warnings | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Every gateway restart shows two identical warnings about openclaw-weixin missing channelConfigs metadata, even when the plugin is disabled.
### Steps to reproduce
1. Install OpenClaw 2026.4.29
2. Set plugins.... | closed | completed | false | 2 | [
"bug"
] | [] | 2026-05-02T13:50:33Z | 2026-05-04T06:39:53Z | 2026-05-04T06:39:53Z | NONE | null | 20260504T090105Z | 2026-05-04T09:01:05Z | googlerest | 127,843,198 | U_kgDOB567fg | User | false |
openclaw/openclaw | 4,369,077,434 | I_kwDOQb6kR88AAAABBGrUug | 76,096 | https://github.com/openclaw/openclaw/issues/76096 | https://api.github.com/repos/openclaw/openclaw/issues/76096 | node.list RPC blocks event loop 30-48s when orphaned agent directories exist | ## Summary
The `node.list` RPC call performs a filesystem scan over **all** directories in `~/.openclaw/agents/` on every invocation, including directories that are orphaned (not present in `agents.list`). On hosts with many stale agent directories and slow storage (eMMC/SD card), this causes the Node.js event loop to... | open | null | false | 1 | [] | [] | 2026-05-02T13:52:03Z | 2026-05-03T11:26:14Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | DBOYttt | 65,898,012 | MDQ6VXNlcjY1ODk4MDEy | User | false |
openclaw/openclaw | 4,368,880,968 | I_kwDOQb6kR88AAAABBGfVSA | 76,071 | https://github.com/openclaw/openclaw/issues/76071 | https://api.github.com/repos/openclaw/openclaw/issues/76071 | [Bug]: Gemini 3.1 Pro Preview hangs in agent runtime; direct API works in seconds, openclaw idle-times-out at full timeoutSeconds | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
google/gemini-3.1-pro-preview always hits LLM idle timeout: no res... | closed | completed | true | 1 | [
"bug"
] | [] | 2026-05-02T12:24:26Z | 2026-05-05T04:27:07Z | 2026-05-02T13:56:56Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | marcoschierhorn | 384,813 | MDQ6VXNlcjM4NDgxMw== | User | false |
openclaw/openclaw | 4,369,082,846 | I_kwDOQb6kR88AAAABBGrp3g | 76,098 | https://github.com/openclaw/openclaw/issues/76098 | https://api.github.com/repos/openclaw/openclaw/issues/76098 | [openclaw-weixin] 微信拍照发送单张图片时，桥接层附带不相关的额外图片 | ## Bug 描述
通过微信聊天的拍照功能发送单张图片时，OpenClaw 微信桥接层会将不相关的额外图片打包进同一条消息，导致 AI agent 收到两张图片。
## 环境
- OpenClaw 版本：2026.4.22
- 通道：openclaw-weixin（iLink 企业接口）
- 消息类型：单张图片（拍照发送，无文字）
## 复现步骤
1. 通过微信聊天界面使用拍照功能，拍摄一张照片并发送给机器人
2. AI agent 收到的消息中包含 **两张图片**，而非一张
## 实际行为
Session 原始记录显示，用户消息中包含两个 image data 条目：
- 图片 1：不相关的旧图片（路由器截图，保存... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T13:53:52Z | 2026-05-05T04:27:06Z | 2026-05-02T13:58:54Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | adcosgou | 170,591,273 | U_kgDOCisEKQ | User | false |
openclaw/openclaw | 4,369,094,097 | I_kwDOQb6kR88AAAABBGsV0Q | 76,101 | https://github.com/openclaw/openclaw/issues/76101 | https://api.github.com/repos/openclaw/openclaw/issues/76101 | Lobster docs show openclaw.invoke --tool llm-task, but embedded runner lacks gateway URL/auth context | ## Summary
The shipped docs show `openclaw.invoke --tool llm-task` as the way to run an `llm-task` step from a Lobster workflow, but the embedded Lobster runner does not appear to provide the gateway URL/auth context needed for that command to work.
On OpenClaw `2026.4.29 (a448042)`, the documented Lobster -> `opencl... | open | null | false | 1 | [] | [] | 2026-05-02T13:58:11Z | 2026-05-03T10:20:27Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | richardmqq | 14,178,267 | MDQ6VXNlcjE0MTc4MjY3 | User | false |
openclaw/openclaw | 4,369,096,523 | I_kwDOQb6kR88AAAABBGsfSw | 76,102 | https://github.com/openclaw/openclaw/issues/76102 | https://api.github.com/repos/openclaw/openclaw/issues/76102 | [Bug]: WhatsApp 408 timeout loop, `channels login` hangs indefinitely, and agent reply failure on v2026.4.29 (Windows) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Upgrading from 2026.4.23 to 2026.4.29 on Windows causes three related failures:
1. **WhatsApp WebSocket drops with 408 Request Time-out** every ~60 seconds, killing connections before the agent can deliver replies
2. **`ope... | closed | not_planned | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T13:59:15Z | 2026-05-05T04:27:05Z | 2026-05-02T14:04:38Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | zaff70 | 272,942,594 | U_kgDOEETGAg | User | false |
openclaw/openclaw | 4,369,111,087 | I_kwDOQb6kR88AAAABBGtYLw | 76,104 | https://github.com/openclaw/openclaw/issues/76104 | https://api.github.com/repos/openclaw/openclaw/issues/76104 | sessions_send reply routes to webchat instead of origin channel (Feishu) | ## Bug Report: sessions_send reply routes to webchat instead of origin channel (Feishu)
### Environment
- **OpenClaw version**: 2026.4.29
- **Channel**: Feishu (websocket mode)
- **Issue occur**: When triggering sessions_send from Feishu DM
### Description
When a user sends a command (e.g., /compact) via Feishu dire... | open | null | false | 1 | [] | [] | 2026-05-02T14:04:54Z | 2026-05-03T11:26:15Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | hongfangsong | 197,858,246 | U_kgDOC8sTxg | User | false |
openclaw/openclaw | 4,369,078,547 | I_kwDOQb6kR88AAAABBGrZEw | 76,097 | https://github.com/openclaw/openclaw/issues/76097 | https://api.github.com/repos/openclaw/openclaw/issues/76097 | Bug: sqlite-vec missing from plugin-runtime-deps — memory indexing completely broken | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
Yes
### Summary
Description
sqlite-vec is not declared in plugin-runtime-deps/package.json, so npm install never installs it. This causes the memory vector indexing to fail silently.
### Steps to reproduce
Symptoms
Gateway ... | closed | completed | true | 2 | [
"bug"
] | [] | 2026-05-02T13:52:28Z | 2026-05-05T04:26:48Z | 2026-05-02T14:12:27Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | a741800221 | 72,013,187 | MDQ6VXNlcjcyMDEzMTg3 | User | false |
openclaw/openclaw | 4,369,137,116 | I_kwDOQb6kR88AAAABBGu93A | 76,108 | https://github.com/openclaw/openclaw/issues/76108 | https://api.github.com/repos/openclaw/openclaw/issues/76108 | Model fallback doesn't recover after cooldown — persists across sessions | ## Bug description
When `minimax-portal/MiniMax-M2.7` triggers a **format error** and enters cooldown, OpenClaw correctly falls back to the next available model (`volcengine-plan/ark-code-latest`). However, **after the cooldown expires**, OpenClaw does NOT automatically recover the primary model. Instead:
1. The sess... | closed | not_planned | false | 1 | [] | [] | 2026-05-02T14:14:36Z | 2026-05-03T11:33:43Z | 2026-05-03T11:33:43Z | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | rongtianhua | 45,085,618 | MDQ6VXNlcjQ1MDg1NjE4 | User | false |
openclaw/openclaw | 4,369,137,403 | I_kwDOQb6kR88AAAABBGu--w | 76,109 | https://github.com/openclaw/openclaw/issues/76109 | https://api.github.com/repos/openclaw/openclaw/issues/76109 | Gateway: Event loop delay grows to 40s+ and CPU pinned at 100% on v2026.4.29 | ### Summary
Gateway process on v2026.4.29 enters a dead-loop state: CPU pinned at 100%, event loop delay exceeding 40s, rendering the service unresponsive to incoming messages across all channels (WeChat, Feishu, Discord).
### Repro Steps
1. Start gateway with `openclaw gateway run` (systemd-managed) on a standard L... | closed | completed | true | 1 | [] | [] | 2026-05-02T14:14:40Z | 2026-05-05T04:27:04Z | 2026-05-02T14:19:10Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | wangyan411303 | 73,583,781 | MDQ6VXNlcjczNTgzNzgx | User | false |
openclaw/openclaw | 4,369,090,114 | I_kwDOQb6kR88AAAABBGsGQg | 76,100 | https://github.com/openclaw/openclaw/issues/76100 | https://api.github.com/repos/openclaw/openclaw/issues/76100 | Race condition: bundled runtime dep json5 missing when Telegram providers start | ## Bug Description
When OpenClaw gateway starts, multiple plugins stage and install their bundled runtime dependencies in parallel. A race condition can cause one plugin (e.g. memory-core, which is smaller and finishes faster) to write its `node_modules` manifest first, and another plugin (e.g. telegram) to then incor... | closed | completed | true | 2 | [] | [] | 2026-05-02T13:56:40Z | 2026-05-05T04:26:48Z | 2026-05-02T14:28:54Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | itsclawdettebot-debug | 260,100,625 | U_kgDOD4DSEQ | User | false |
openclaw/openclaw | 4,369,103,165 | I_kwDOQb6kR88AAAABBGs5PQ | 76,103 | https://github.com/openclaw/openclaw/issues/76103 | https://api.github.com/repos/openclaw/openclaw/issues/76103 | [Bug]: Custom-named providers reusing bundled api adapters (Azure Foundry) silently lose models on 2026.4.29 | ## Summary
Custom-named provider entries in `models.providers` that reuse bundled `api` adapters (e.g. `openai-responses`, `openai-completions`, `anthropic-messages`) silently stop returning models after upgrading from `2026.4.26` → `2026.4.29`. Rolling back to `2026.4.26` restores them with no other changes.
The pat... | closed | completed | true | 1 | [] | [] | 2026-05-02T14:01:56Z | 2026-05-05T04:27:03Z | 2026-05-02T14:28:59Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | rtkchahal | 198,046,571 | U_kgDOC83zaw | User | false |
openclaw/openclaw | 4,369,131,982 | I_kwDOQb6kR88AAAABBGupzg | 76,107 | https://github.com/openclaw/openclaw/issues/76107 | https://api.github.com/repos/openclaw/openclaw/issues/76107 | Message pipeline has two major performance bottlenecks (auth 6.2s, fixed overhead 16s) | ## Bug Report: Message pipeline has two major performance bottlenecks (auth 6.2s, fixed overhead 16s)
### Environment
- **OpenClaw version**: 2026.4.29
- **Model**: MiniMax-M2.7
- **Network**: WiFi (GPON) — baseline 0.1-0.14s Feishu API latency
### Description
During message pipeline profiling, two major bottlenecks... | closed | completed | true | 4 | [] | [
"BunsDev"
] | 2026-05-02T14:12:48Z | 2026-05-05T04:26:00Z | 2026-05-02T23:39:40Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | hongfangsong | 197,858,246 | U_kgDOC8sTxg | User | false |
openclaw/openclaw | 4,369,190,919 | I_kwDOQb6kR88AAAABBGyQBw | 76,112 | https://github.com/openclaw/openclaw/issues/76112 | https://api.github.com/repos/openclaw/openclaw/issues/76112 | Bug: WhatsApp auto-reply delivery double-fires (deliverWebReply called 2x for same correlationId via dispatchReplyWithBufferedBlockDispatcher) | ## Bug Description
**Determinístico double-fire** of `deliverWebReply` in WhatsApp plugin: every outbound auto-reply is sent **twice** to the same recipient with the same `correlationId`, `runId`, and `connectionId` within 5–21 ms. 100% of recent auto-replies in DM (1:1) sessions are duplicated.
This is distinct from... | closed | completed | false | 1 | [] | [] | 2026-05-02T14:38:47Z | 2026-05-04T09:02:53Z | 2026-05-04T09:02:53Z | NONE | null | 20260504T180039Z | 2026-05-04T18:00:39Z | vasconcelosGilson | 264,295,910 | U_kgDOD8DV5g | User | false |
openclaw/openclaw | 4,368,999,036 | I_kwDOQb6kR88AAAABBGmifA | 76,081 | https://github.com/openclaw/openclaw/issues/76081 | https://api.github.com/repos/openclaw/openclaw/issues/76081 | [Bug]: Discord DM replies much slower after 2026.4.29 upgrade | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading to OpenClaw 2026.4.29, Discord DM replies are noticeably slower, often taking about 30-45 seconds before a response appears.
### Steps to reproduce
1. Run OpenClaw 2026.4.29 with Discord DM enabled on the a... | closed | completed | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T13:21:58Z | 2026-05-05T04:27:01Z | 2026-05-02T14:50:03Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | ctreman-maker | 278,375,956 | U_kgDOEJeuFA | User | false |
openclaw/openclaw | 4,369,209,198 | I_kwDOQb6kR88AAAABBGzXbg | 76,117 | https://github.com/openclaw/openclaw/issues/76117 | https://api.github.com/repos/openclaw/openclaw/issues/76117 | [Bug]: The local ollama llm was invalid after 4.26 version | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
When I updated the vsesion to v2026.4.29, the local ollama LLM was ruined. Before v2026.4.26, the local LLM thinks slowly, but it is precious call the tools, knowing your purpose. And now, the local LLM don't know my command... | closed | completed | false | 2 | [
"bug",
"regression"
] | [] | 2026-05-02T14:47:34Z | 2026-05-03T11:47:38Z | 2026-05-03T11:47:38Z | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | MarsCNS | 13,167,160 | MDQ6VXNlcjEzMTY3MTYw | User | false |
openclaw/openclaw | 4,369,250,021 | I_kwDOQb6kR88AAAABBG125Q | 76,124 | https://github.com/openclaw/openclaw/issues/76124 | https://api.github.com/repos/openclaw/openclaw/issues/76124 | [Bug] Edge TTS reads SSML/XML markup as text instead of speaking content | ## Describe the bug
Edge TTS 在发送语音时，把 SSML/XML 标记语言当作文本朗读出来，而不是只朗读标记内的文字内容。
## Steps to reproduce
1. 配置飞书通道 + Edge TTS provider
2. 飞书群 @机器人 发送文字
3. 机器人回复语音气泡
4. 播放语音 → 听到的是 SSML 源码而非回复文字
## Expected behavior
只朗读 SSML `<speak>` 标签内的文字内容
## Actual behavior
朗读了完整的 SSML/XML 标记，Whisper 转录前 15 秒：
```
[00:00] Speak ve... | open | null | false | 1 | [] | [] | 2026-05-02T15:06:04Z | 2026-05-03T01:23:47Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | 17329971 | 83,849,762 | MDQ6VXNlcjgzODQ5NzYy | User | false |
openclaw/openclaw | 4,369,273,021 | I_kwDOQb6kR88AAAABBG3QvQ | 76,127 | https://github.com/openclaw/openclaw/issues/76127 | https://api.github.com/repos/openclaw/openclaw/issues/76127 | Codex harness can leave TUI/session stuck after app-server client closes | ## Summary
When using the native Codex harness globally (`agentRuntime.id: "codex"`), OpenClaw can get into a state where:
- TUI turns remain active/stuck for several minutes after `prompt.submitted`.
- Diagnostics repeatedly skip recovery because an `active_embedded_run` is present.
- The Codex app-server client clo... | open | null | false | 1 | [] | [] | 2026-05-02T15:14:28Z | 2026-05-03T01:23:48Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | bdjben | 188,106,718 | U_kgDOCzZH3g | User | false |
openclaw/openclaw | 4,369,281,286 | I_kwDOQb6kR88AAAABBG3xBg | 76,128 | https://github.com/openclaw/openclaw/issues/76128 | https://api.github.com/repos/openclaw/openclaw/issues/76128 | models set leaves main/default agent pinned to old model via agents.list override | ## Summary
Changing the default model to `openai-codex/gpt-5.5` can appear successful while the active main/default agent continues using the previous model because `agents.list[].id == "main"` has its own explicit `model.primary`.
In my case, the global default had been updated to 5.5, but the active `main`/Jinx pat... | open | null | false | 2 | [] | [] | 2026-05-02T15:18:18Z | 2026-05-03T06:51:09Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | trialanderrorstudios | 273,140,757 | U_kgDOEEfMFQ | User | false |
openclaw/openclaw | 4,369,290,194 | I_kwDOQb6kR88AAAABBG4T0g | 76,130 | https://github.com/openclaw/openclaw/issues/76130 | https://api.github.com/repos/openclaw/openclaw/issues/76130 | [Bug]: Telegram CLI message send still hangs on 2026.4.29 after grammy runtime-deps fix | ## Bug type
Regression / Telegram CLI send hang / runtime dependency resolution
## Related issue
This follows up on #73140.
I tried to comment there, but GitHub returned:
```text
GraphQL: Unable to create comment because issue is locked (addComment)
```
## Summary
`openclaw message send --channel telegram ... --... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T15:22:45Z | 2026-05-05T04:26:58Z | 2026-05-02T15:27:14Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | oalansilva | 126,212 | MDQ6VXNlcjEyNjIxMg== | User | false |
openclaw/openclaw | 4,369,319,559 | I_kwDOQb6kR88AAAABBG6Ghw | 76,138 | https://github.com/openclaw/openclaw/issues/76138 | https://api.github.com/repos/openclaw/openclaw/issues/76138 | [Feature]: New Session "+" button: add confirmation, change icon, or make it configurable | ### Summary
Prevent accidental session loss from the New Session button in the chat toolbar
### Problem to solve
The "+" (New Session) button sits in the chat input toolbar right next to the Send button. A single misclick triggers /new, which immediately replaces the current conversation with a blank session — with ... | open | null | false | 4 | [
"enhancement"
] | [] | 2026-05-02T15:35:56Z | 2026-05-03T20:25:17Z | null | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | ooecoo | 56,377,631 | MDQ6VXNlcjU2Mzc3NjMx | User | false |
openclaw/openclaw | 4,369,307,433 | I_kwDOQb6kR88AAAABBG5XKQ | 76,135 | https://github.com/openclaw/openclaw/issues/76135 | https://api.github.com/repos/openclaw/openclaw/issues/76135 | Allow private messaging group sessions to surface verbose/tool progress | ## Summary
Signal group chats (and likely other non-forum messaging group/channel surfaces) can be used as durable, private per-project OpenClaw sessions, but they currently cannot fully simulate the main/direct chat experience because verbose/tool progress output is suppressed for `ChatType === "group"`.
This makes ... | open | null | false | 1 | [] | [] | 2026-05-02T15:30:24Z | 2026-05-03T06:37:02Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | arduano | 13,347,712 | MDQ6VXNlcjEzMzQ3NzEy | User | false |
openclaw/openclaw | 4,369,376,180 | I_kwDOQb6kR88AAAABBG9jtA | 76,144 | https://github.com/openclaw/openclaw/issues/76144 | https://api.github.com/repos/openclaw/openclaw/issues/76144 | Plugin loader rejects node-owned plugin dirs with misleading "stale config entry" surrounding warnings; container restart auto-reverts manual chown | ## Problem
On openclaw 2026.4.12, the gateway plugin loader rejects any plugin whose source directory is owned by the `node` user (uid 1000) with:
```
[gateway] [plugins] blocked plugin candidate: suspicious ownership
(/data/plugins/<plugin-name>, uid=1000, expected uid=0 or root)
```
This is reasonable defense-i... | closed | completed | false | 1 | [] | [] | 2026-05-02T16:01:18Z | 2026-05-03T18:39:31Z | 2026-05-03T18:39:31Z | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | mayank6136 | 16,468,779 | MDQ6VXNlcjE2NDY4Nzc5 | User | false |
openclaw/openclaw | 4,369,376,204 | I_kwDOQb6kR88AAAABBG9jzA | 76,145 | https://github.com/openclaw/openclaw/issues/76145 | https://api.github.com/repos/openclaw/openclaw/issues/76145 | `openclaw agent` CLI exits before parallel sub-agents finish; reported `finalAssistantVisibleText` only captures the parent's first turn | ## Problem
When the parent agent uses `sessions_spawn` to engage sub-agents in parallel, the CLI's foreground process exits as soon as the parent emits its first text turn (the "announce" turn — e.g., "I've engaged telemetry, backend, and db. Waiting on findings…"), even though:
1. Sub-agents are still mid-LLM-call w... | open | null | false | 2 | [] | [] | 2026-05-02T16:01:18Z | 2026-05-03T04:44:45Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | mayank6136 | 16,468,779 | MDQ6VXNlcjE2NDY4Nzc5 | User | false |
openclaw/openclaw | 4,369,398,464 | I_kwDOQb6kR88AAAABBG-6wA | 76,147 | https://github.com/openclaw/openclaw/issues/76147 | https://api.github.com/repos/openclaw/openclaw/issues/76147 | `subagent-registry.runtime.js` still missing from `dist/` in 2026.4.12 npm package — file genuinely absent (follow-up to closed #71418) | ## Symptom
On a fresh gateway boot of openclaw 2026.4.12, the following warning fires 18 times:
```
[warn] subagent cleanup finalize failed (<uuid>): Error [ERR_MODULE_NOT_FOUND]:
Cannot find module '/data/.npm-global/lib/node_modules/openclaw/dist/subagent-registry.runtime.js'
imported from /data/.npm-global/li... | closed | completed | true | 1 | [] | [] | 2026-05-02T16:09:21Z | 2026-05-05T04:26:51Z | 2026-05-02T16:13:36Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | mayank6136 | 16,468,779 | MDQ6VXNlcjE2NDY4Nzc5 | User | false |
openclaw/openclaw | 4,369,407,591 | I_kwDOQb6kR88AAAABBG_eZw | 76,149 | https://github.com/openclaw/openclaw/issues/76149 | https://api.github.com/repos/openclaw/openclaw/issues/76149 | [Feature]: Telegram — expose mediaGroupTimeoutMs / mediaGroupFlushMs as a user-facing config key (follow-up to #1811) | ## Problem
The Telegram extension's media-group buffer flush window is hardcoded to 500ms and is only overridable via `opts.testTimings.mediaGroupFlushMs`, which is a test-only hook with no user-facing config path.
When network latency between Telegram's CDN and the gateway causes some album items to arrive >500ms af... | closed | completed | false | 2 | [] | [
"vincentkoc"
] | 2026-05-02T16:12:59Z | 2026-05-03T19:13:27Z | 2026-05-03T19:13:27Z | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | Chris2Eng | 30,375,456 | MDQ6VXNlcjMwMzc1NDU2 | User | false |
openclaw/openclaw | 4,369,407,677 | I_kwDOQb6kR88AAAABBG_evQ | 76,150 | https://github.com/openclaw/openclaw/issues/76150 | https://api.github.com/repos/openclaw/openclaw/issues/76150 | [Bug]: Gateway did not become healthy after restart. | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Gateway version mismatch: expected 2026.4.29, running gateway reported unavailable.
Service runtime: status=running, state=active, pid=1675, lastExit=0
Gateway port 18789 status: free.
Port diagnostics errors: Error: spawn l... | open | null | false | 1 | [
"bug"
] | [] | 2026-05-02T16:13:01Z | 2026-05-04T01:47:50Z | null | NONE | null | 20260504T030031Z | 2026-05-04T03:00:31Z | tzzym | 96,649,677 | U_kgDOBcLBzQ | User | false |
openclaw/openclaw | 4,369,295,634 | I_kwDOQb6kR88AAAABBG4pEg | 76,132 | https://github.com/openclaw/openclaw/issues/76132 | https://api.github.com/repos/openclaw/openclaw/issues/76132 | [Bug]: Onboarding Bug - Stuck at Hatching | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Agent hatching is stuck. Agent does not wake up. The TUI chat is frozen, not responding to key strokes.
### Steps to reproduce
OS: Ubuntu 26
LLM Provider: OpenAI Cloud
Found in Version 2026.4.29
Node Version: tried on ... | closed | completed | true | 3 | [
"bug",
"regression"
] | [
"joshavant"
] | 2026-05-02T15:25:25Z | 2026-05-05T04:26:21Z | 2026-05-02T20:28:52Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | rossbrigoli | 1,734,075 | MDQ6VXNlcjE3MzQwNzU= | User | false |
openclaw/openclaw | 4,369,457,507 | I_kwDOQb6kR88AAAABBHChYw | 76,158 | https://github.com/openclaw/openclaw/issues/76158 | https://api.github.com/repos/openclaw/openclaw/issues/76158 | [Bug]: BlueBubbles tapback events reach before_dispatch as synthesized text bodies; structural fields (associatedMessageType / isTapback) are stripped, enabling empty-turn synthetic leak | ## Summary
The BlueBubbles channel plugin parses inbound tapback / reaction events into structured form (`resolveTapbackContext`, `isTapbackAssociatedType`, `associatedMessageType` 2000–2999 = added / 3000–3999 = removed), but the structural fields are **not propagated into the `before_dispatch` plugin hook event**. P... | closed | completed | true | 1 | [] | [] | 2026-05-02T16:36:48Z | 2026-05-05T04:26:45Z | 2026-05-02T16:41:16Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | mids-neo | 117,532,760 | U_kgDOBwFoWA | User | false |
openclaw/openclaw | 4,369,437,173 | I_kwDOQb6kR88AAAABBHBR9Q | 76,154 | https://github.com/openclaw/openclaw/issues/76154 | https://api.github.com/repos/openclaw/openclaw/issues/76154 | [Bug]: TUI enters infinite CPU loop immediately on launch (v2026.4.29) | ### Bug Summary
`openclaw tui` (and aliases `chat`, `terminal`) enter an infinite CPU busy-loop immediately upon launch. The TUI process never becomes interactive and consumes 100-115% CPU within seconds.
### Impact
- Cannot use terminal UI at all
- Workaround: webchat dashboard works fine
### Environment
| Item | Va... | open | null | false | 1 | [] | [] | 2026-05-02T16:26:51Z | 2026-05-03T03:38:21Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | pedrof123-ui | 249,927,142 | U_kgDODuWV5g | User | false |
openclaw/openclaw | 4,369,440,062 | I_kwDOQb6kR88AAAABBHBdPg | 76,155 | https://github.com/openclaw/openclaw/issues/76155 | https://api.github.com/repos/openclaw/openclaw/issues/76155 | [Bug] cron-nested lane mis-resolves hyphen-prefixed provider aliases on 2026.4.29 | # [Bug] cron-nested lane mis-resolves hyphen-prefixed provider aliases on 2026.4.29
## Summary
The `cron-nested` lane's model resolver mis-handles aliases of the form `<provider-with-hyphen>/<model>` (e.g., `kimi-code/kimi-for-coding`), producing a stripped/different alias and reporting `Unknown model`. The same alia... | open | null | false | 1 | [] | [] | 2026-05-02T16:28:16Z | 2026-05-03T03:38:22Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | chrislro | 7,277,502 | MDQ6VXNlcjcyNzc1MDI= | User | false |
openclaw/openclaw | 4,369,468,997 | I_kwDOQb6kR88AAAABBHDORQ | 76,160 | https://github.com/openclaw/openclaw/issues/76160 | https://api.github.com/repos/openclaw/openclaw/issues/76160 | cron: cron add/edit silently allows duplicate name collisions | ## Summary
`openclaw cron add` (and `cron edit`) silently accept new jobs whose `name` collides with an existing job's `name`. Two distinct rows with identical names land in `jobs.json`, each with its own `id`, `schedule`, `enabled`, etc. Default `cron list` only renders one of them when their `enabled` differ, hiding... | open | null | false | 1 | [] | [] | 2026-05-02T16:41:43Z | 2026-05-03T03:38:24Z | null | CONTRIBUTOR | null | 20260503T060030Z | 2026-05-03T06:00:30Z | markthebest12 | 6,416,874 | MDQ6VXNlcjY0MTY4NzQ= | User | false |
openclaw/openclaw | 4,369,462,285 | I_kwDOQb6kR88AAAABBHC0DQ | 76,159 | https://github.com/openclaw/openclaw/issues/76159 | https://api.github.com/repos/openclaw/openclaw/issues/76159 | [Feature]: Per-job acceptSilentStop flag — treat stopReason=stop with payloads=0 as ok for cron jobs that intentionally produce no output | ## Problem
Cron jobs whose prompts say "if no work to do, output nothing" are marked as
errors by the framework. The job runs fine, the model decides correctly that
no action is needed, but because `stopReason=stop` with `payloads.length=0`
the run is classified as an "incomplete turn" — `lastError: ⚠️ Agent
couldn't ... | open | null | false | 1 | [] | [] | 2026-05-02T16:38:40Z | 2026-05-03T22:24:14Z | null | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | RonnyMarkus | 44,967,939 | MDQ6VXNlcjQ0OTY3OTM5 | User | false |
openclaw/openclaw | 4,369,494,064 | I_kwDOQb6kR88AAAABBHEwMA | 76,162 | https://github.com/openclaw/openclaw/issues/76162 | https://api.github.com/repos/openclaw/openclaw/issues/76162 | [Bug] exec-approval-followup tasks stuck in 'running' forever, blocking channel reload and saturating event loop | ## Summary
When exec approval follow-up tasks fail (e.g. network error during LLM call), the task runs remain in `running` status indefinitely. They are not reconciled by the task maintenance sweeper because `hasBackingSession` always returns `true` for `cli` runtime tasks whose `childSessionKey` points to a persisten... | open | null | false | 2 | [] | [] | 2026-05-02T16:53:30Z | 2026-05-03T11:34:12Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | DuanXiaoWen | 64,217,807 | MDQ6VXNlcjY0MjE3ODA3 | User | false |
openclaw/openclaw | 4,369,297,878 | I_kwDOQb6kR88AAAABBG4x1g | 76,133 | https://github.com/openclaw/openclaw/issues/76133 | https://api.github.com/repos/openclaw/openclaw/issues/76133 | [Bug]: Telegram stop messages after upgrade to openclaw upgrade (2026.04.29) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from 2026.04.26 to 2026.04.29, telegram integration stopped.
The messa is not shown anymore:
15:05:37 [heartbeat] started
15:05:38 [telegram] [default] starting provider (@openclaw_ai_nia_bot)
After downgrad... | closed | not_planned | true | 2 | [
"bug",
"regression"
] | [] | 2026-05-02T15:26:30Z | 2026-05-05T04:26:44Z | 2026-05-02T15:35:27Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Mosheh | 191,274 | MDQ6VXNlcjE5MTI3NA== | User | false |
openclaw/openclaw | 4,369,515,066 | I_kwDOQb6kR88AAAABBHGCOg | 76,167 | https://github.com/openclaw/openclaw/issues/76167 | https://api.github.com/repos/openclaw/openclaw/issues/76167 | Windows ESM regression: Telegram channel loader passes raw C:\ paths to strict ESM loader (v2026.4.24) | ### Bug type
Crash / fatal error
### Beta release blocker
No
### Summary
Split from #75687 (closed prematurely — only Bugs 1–2 were addressed on `main`). This is **Bug 5** from that report.
On Windows, the Telegram channel loader passes raw `C:\...` file paths to Node 22's strict ESM loader, which rejects them:
... | closed | completed | true | 1 | [] | [] | 2026-05-02T17:03:18Z | 2026-05-05T04:26:42Z | 2026-05-02T17:07:12Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | vdruts | 20,264,016 | MDQ6VXNlcjIwMjY0MDE2 | User | false |
openclaw/openclaw | 4,369,518,725 | I_kwDOQb6kR88AAAABBHGQhQ | 76,168 | https://github.com/openclaw/openclaw/issues/76168 | https://api.github.com/repos/openclaw/openclaw/issues/76168 | [BUG] message send CLI can hang as openclaw-message at ~100% CPU after plugin load failure | ## Summary
On OpenClaw `2026.4.29`, `openclaw message send` can spawn a child process titled `openclaw-message` that gets stuck at ~100% CPU instead of exiting after a Telegram plugin load failure. While that process is spinning, the gateway becomes unhealthy (very large event-loop delays, Telegram replies stall, user... | closed | completed | false | 2 | [] | [] | 2026-05-02T17:04:58Z | 2026-05-03T17:32:38Z | 2026-05-03T17:32:38Z | NONE | null | 20260503T180029Z | 2026-05-03T18:00:29Z | waxdr57913 | 105,788,350 | U_kgDOBk4zvg | User | false |
openclaw/openclaw | 4,369,526,396 | I_kwDOQb6kR88AAAABBHGufA | 76,170 | https://github.com/openclaw/openclaw/issues/76170 | https://api.github.com/repos/openclaw/openclaw/issues/76170 | Web UI blank screen after login on 2026.4.30-beta.1 | ## Description
After upgrading from `2026.4.29` to `2026.4.30-beta.1`, the web dashboard shows a blank screen immediately after entering the password. Hard refresh and incognito window both reproduce the issue.
## Environment
- **Version:** 2026.4.30-beta.1
- **Deployment:** Railway (Docker, node:24-bookworm-slim)
- ... | closed | completed | true | 4 | [] | [] | 2026-05-02T17:08:25Z | 2026-05-05T04:25:56Z | 2026-05-02T21:36:27Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Vianne-droid | 265,823,039 | U_kgDOD9gjPw | User | false |
openclaw/openclaw | 4,369,534,187 | I_kwDOQb6kR88AAAABBHHM6w | 76,172 | https://github.com/openclaw/openclaw/issues/76172 | https://api.github.com/repos/openclaw/openclaw/issues/76172 | [Bug]: [Telegram] sendChatAction and sendMessage fail with "Network request failed" — polling works, direct grammy/undici calls succeed | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
The Telegram channel fails to send any outbound messages. Both sendChatAction and sendMessage consistently fail with:
Network request for 'sendChatAction' failed!
Network request for 'sendChatAction' failed!
T... | closed | not_planned | true | 2 | [
"bug",
"bug:behavior"
] | [] | 2026-05-02T17:12:22Z | 2026-05-05T04:26:14Z | 2026-05-02T17:17:56Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | nathanrhoffman-cmd | 265,971,606 | U_kgDOD9pnlg | User | false |
openclaw/openclaw | 4,369,546,322 | I_kwDOQb6kR88AAAABBHH8Ug | 76,174 | https://github.com/openclaw/openclaw/issues/76174 | https://api.github.com/repos/openclaw/openclaw/issues/76174 | [Bug]: 2026.4.29 — all openai/* embedded runs hang with zero tokens until timeout; direct curl to OpenAI works in 2s | ## Summary
On OpenClaw 2026.4.29, **every `openai/*` model embedded run hangs for the full timeout window with zero tokens emitted** and ends with `FailoverError: LLM request timed out.` Direct `curl` from the same host to `api.openai.com` using the same API key returns a real response in ~1–2 seconds, so the OpenAI P... | closed | not_planned | true | 6 | [] | [] | 2026-05-02T17:18:17Z | 2026-05-05T04:25:19Z | 2026-05-03T02:04:21Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | pamopsdev | 281,040,568 | U_kgDOEMBWuA | User | false |
openclaw/openclaw | 4,369,505,095 | I_kwDOQb6kR88AAAABBHFbRw | 76,164 | https://github.com/openclaw/openclaw/issues/76164 | https://api.github.com/repos/openclaw/openclaw/issues/76164 | Telegram polling stalls and sendMessage failures on Windows multi-bot (v2026.4.24–.25) | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Split from #75687 (closed prematurely — only Bugs 1–2 were addressed on `main`). This is **Bug 3** from that report.
On Windows with 8 Telegram bots, versions `.24` and `.25` exhibit Telegram polling stalls an... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T16:58:40Z | 2026-05-05T04:26:42Z | 2026-05-02T17:25:03Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | vdruts | 20,264,016 | MDQ6VXNlcjIwMjY0MDE2 | User | false |
openclaw/openclaw | 4,369,510,003 | I_kwDOQb6kR88AAAABBHFucw | 76,165 | https://github.com/openclaw/openclaw/issues/76165 | https://api.github.com/repos/openclaw/openclaw/issues/76165 | Model fallback cascade takes 30s per invalid candidate instead of failing fast (v2026.4.24–.25) | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Split from #75687 (closed prematurely — only Bugs 1–2 were addressed on `main`). This is **Bug 4** from that report.
Non-retryable `400` errors (e.g. invalid model ID, unsupported parameter) take ~30 seconds e... | open | null | false | 1 | [] | [] | 2026-05-02T17:01:01Z | 2026-05-03T22:24:15Z | null | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | vdruts | 20,264,016 | MDQ6VXNlcjIwMjY0MDE2 | User | false |
openclaw/openclaw | 4,369,510,539 | I_kwDOQb6kR88AAAABBHFwiw | 76,166 | https://github.com/openclaw/openclaw/issues/76166 | https://api.github.com/repos/openclaw/openclaw/issues/76166 | Control UI still repeatedly calls slow sessions.list on 2026.4.29, causing Gateway latency/GC pressure | ## Summary
On OpenClaw `2026.4.29 (a448042)`, the local Control UI/webchat connection still repeatedly called `sessions.list`, and each call took about 10–11s on a large session store. This caused Gateway event-loop/GC pressure and user-visible RPC timeouts.
This looks like a regression or release-build mismatch with... | closed | completed | true | 4 | [] | [
"BunsDev"
] | 2026-05-02T17:01:16Z | 2026-05-05T04:25:59Z | 2026-05-02T23:39:42Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | richardmqq | 14,178,267 | MDQ6VXNlcjE0MTc4MjY3 | User | false |
openclaw/openclaw | 4,369,523,713 | I_kwDOQb6kR88AAAABBHGkAQ | 76,169 | https://github.com/openclaw/openclaw/issues/76169 | https://api.github.com/repos/openclaw/openclaw/issues/76169 | openai-completions chat requests bypass HTTPS_PROXY in proxy-required sandbox (regression of closed #56841) | ### Environment
- OpenClaw `2026.4.26` (also reproduced on `2026.4.23`)
- Pi `0.70`
- Node `v22.22.1`
- Bundled openai SDK `v6.35.0`
- DGX Spark (GB10) running NVIDIA openshell community image, sandbox connectivity gated by `HTTPS_PROXY=http://127.0.0.1:3128 → 10.200.0.1:3128 → openshell egress proxy`
- All non-loopba... | closed | completed | false | 1 | [] | [] | 2026-05-02T17:07:12Z | 2026-05-03T21:04:55Z | 2026-05-03T21:04:55Z | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | HeilbronAILabs | 274,369,335 | U_kgDOEFqLNw | User | false |
openclaw/openclaw | 4,369,533,615 | I_kwDOQb6kR88AAAABBHHKrw | 76,171 | https://github.com/openclaw/openclaw/issues/76171 | https://api.github.com/repos/openclaw/openclaw/issues/76171 | High host load & slow responses caused by stale openclaw worker process accumulation | ## Description
On `2026.4.29` (and persisting into `2026.4.30-beta.1`), response times degrade to 2-3+ minutes per turn due to stale openclaw worker processes accumulating on the host and driving load average to 25-31.
## Environment
- **Version:** 2026.4.29 / 2026.4.30-beta.1
- **Deployment:** Railway (Docker, node:... | open | null | false | 3 | [] | [] | 2026-05-02T17:12:04Z | 2026-05-03T06:08:28Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | Vianne-droid | 265,823,039 | U_kgDOD9gjPw | User | false |
openclaw/openclaw | 4,369,536,054 | I_kwDOQb6kR88AAAABBHHUNg | 76,173 | https://github.com/openclaw/openclaw/issues/76173 | https://api.github.com/repos/openclaw/openclaw/issues/76173 | [Feature]: Simplify the dashboard UX | ### Summary
Make the dashboard simpler, cleaner, and easier to use.
### Problem to solve
The current dashboard feels too complex and can be difficult to navigate, especially for new users. Important actions and information are not always easy to find quickly, which slows down daily usage and creates friction.
A sim... | closed | not_planned | true | 3 | [
"enhancement"
] | [] | 2026-05-02T17:13:16Z | 2026-05-05T04:25:31Z | 2026-05-02T21:23:32Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | markdudov | 126,427,334 | U_kgDOB4kgxg | User | false |
openclaw/openclaw | 4,369,242,910 | I_kwDOQb6kR88AAAABBG1bHg | 76,123 | https://github.com/openclaw/openclaw/issues/76123 | https://api.github.com/repos/openclaw/openclaw/issues/76123 | [Bug]: 2026.4.29 performance regression — +43% request latency, stuck sessions, event loop blocking | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from 2026.4.26 to 2026.4.29 (Pi framework 0.70.6), all request latency increased 35-40% (~75s → ~106s), with 124 liveness warnings (P99 peak 48s), stuck sessions lasting 10+ minutes, 36 failover events, and 5... | closed | completed | false | 11 | [
"bug",
"regression"
] | [] | 2026-05-02T15:03:10Z | 2026-05-02T20:21:48Z | 2026-05-02T20:20:36Z | NONE | null | 20260502T210030Z | 2026-05-02T21:00:30Z | linguoweipost | 281,120,969 | U_kgDOEMGQyQ | User | false |
openclaw/openclaw | 4,369,580,372 | I_kwDOQb6kR88AAAABBHKBVA | 76,178 | https://github.com/openclaw/openclaw/issues/76178 | https://api.github.com/repos/openclaw/openclaw/issues/76178 | Add Russian locale to Control UI | ## Summary
I've created a complete Russian translation for the OpenClaw Control UI dashboard.
## Changes
1. **`ui/src/i18n/locales/ru.ts`** — Full Russian translation (1200+ strings covering all UI sections)
2. **`ui/src/i18n/lib/types.ts`** — Added `"ru"` to the `Locale` type union
3. **`ui/src/i18n/lib/registry.ts... | closed | not_planned | true | 2 | [] | [] | 2026-05-02T17:33:15Z | 2026-05-05T04:26:40Z | 2026-05-02T17:36:49Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | warment | 238,870,369 | U_kgDODjzfYQ | User | false |
openclaw/openclaw | 4,369,590,343 | I_kwDOQb6kR88AAAABBHKoRw | 76,182 | https://github.com/openclaw/openclaw/issues/76182 | https://api.github.com/repos/openclaw/openclaw/issues/76182 | [Bug]: severe main-thread stalls from repeated plugin-runtime-deps manifest scanning (plugin-runtime-deps tree + repeated openclaw.plugin.json rereads) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from 2026.3.13 to 2026.4.29, the gateway becomes effectively unusable on a Pi 4 due to severe event-loop stalls and main command lane starvation; the regression appears tied to the plugin-runtime-deps tree an... | closed | completed | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T17:37:15Z | 2026-05-05T04:26:39Z | 2026-05-02T17:42:01Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | github-vanity | 76,969,728 | MDQ6VXNlcjc2OTY5NzI4 | User | false |
openclaw/openclaw | 4,369,600,411 | I_kwDOQb6kR88AAAABBHLPmw | 76,187 | https://github.com/openclaw/openclaw/issues/76187 | https://api.github.com/repos/openclaw/openclaw/issues/76187 | [Bug]: Gateway hangs at plugins.bootstrap on 2026.4.26 (never reaches ready) | ## Summary
On 2026.4.26 image, `openclaw gateway` hangs **before reaching `ready`**, with no log output after `starting...`. The hang is inside `prepareGatewayPluginBootstrap` — the next-expected `startupTrace.measure("plugins.bootstrap", ...)` call never completes, so port 18789 is never listened on and `/healthz` is... | closed | completed | true | 1 | [] | [] | 2026-05-02T17:42:10Z | 2026-05-05T04:26:31Z | 2026-05-02T19:11:40Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | jeffchen1981-fu | 260,034,475 | U_kgDOD3_Pqw | User | false |
openclaw/openclaw | 4,369,598,480 | I_kwDOQb6kR88AAAABBHLIEA | 76,186 | https://github.com/openclaw/openclaw/issues/76186 | https://api.github.com/repos/openclaw/openclaw/issues/76186 | [Feature]: Session prep phase (~48s) re-executes on EVERY turn, not cached within session | ### Summary
Every user message within the same session triggers a full rebuild of the prep pipeline costing ~48-50 seconds before the model even starts generating. `system-prompt`, `core-plugin-tools`, and `stream-setup` are re-executed from scratch on every turn, with no session-level caching.
### Problem to solve
... | closed | completed | false | 5 | [
"enhancement"
] | [] | 2026-05-02T17:41:12Z | 2026-05-03T16:25:37Z | 2026-05-03T16:25:37Z | NONE | null | 20260503T180029Z | 2026-05-03T18:00:29Z | tpanda09 | 104,880,376 | U_kgDOBkBY-A | User | false |
openclaw/openclaw | 4,369,600,766 | I_kwDOQb6kR88AAAABBHLQ_g | 76,188 | https://github.com/openclaw/openclaw/issues/76188 | https://api.github.com/repos/openclaw/openclaw/issues/76188 | Performance: chain of issues causing excessive CPU/event-loop saturation on low-power hosts | ## Overview
This issue documents a chain of compounding performance problems that collectively make OpenClaw unusable on low-power hardware (Pi, NUC, homelab mini-PCs). Each issue is independently harmful, but together they cause full CPU saturation and event-loop delays of 30+ seconds.
Related to: #76096 (`node.list... | closed | completed | true | 5 | [] | [
"BunsDev"
] | 2026-05-02T17:42:21Z | 2026-05-05T04:26:01Z | 2026-05-02T23:39:39Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | DBOYttt | 65,898,012 | MDQ6VXNlcjY1ODk4MDEy | User | false |
openclaw/openclaw | 4,369,566,121 | I_kwDOQb6kR88AAAABBHJJqQ | 76,176 | https://github.com/openclaw/openclaw/issues/76176 | https://api.github.com/repos/openclaw/openclaw/issues/76176 | [Bug]: Telegram agent hangs with openai/gpt-5.4-mini; chat-completions fallback sends unsupported reasoning_effort | ### Bug type
Crash (process/app exits or hangs)
### Beta release blocker
No
### Summary
[Bug]: Telegram agent hangs with openai/gpt-5.4-mini; chat-completions fallback sends unsupported reasoning_effort
Environment:
- OpenClaw: 2026.4.29
- macOS: 26.3.1 arm64
- Node: 23.11.0
- Gateway: LaunchAgent, local loopback... | closed | completed | false | 1 | [
"bug",
"bug:crash"
] | [] | 2026-05-02T17:27:38Z | 2026-05-03T15:02:17Z | 2026-05-03T15:02:17Z | NONE | null | 20260503T180029Z | 2026-05-03T18:00:29Z | ThisIsAdilah | 105,906,928 | U_kgDOBlAC8A | User | false |
openclaw/openclaw | 4,369,583,120 | I_kwDOQb6kR88AAAABBHKMEA | 76,179 | https://github.com/openclaw/openclaw/issues/76179 | https://api.github.com/repos/openclaw/openclaw/issues/76179 | [Bug]: Codex app-server frequently restarts native threads when owner-only tool availability flips | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
A single long-lived OpenClaw session using `agentRuntime.id = "codex"` repeatedly starts new Codex app-server native threads when the dynamic tool catalog changes only because per-turn owner-only tool availabil... | open | null | false | 2 | [] | [] | 2026-05-02T17:34:19Z | 2026-05-03T06:37:39Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | songshikang0111 | 132,348,332 | U_kgDOB-N5rA | User | false |
openclaw/openclaw | 4,369,610,876 | I_kwDOQb6kR88AAAABBHL4fA | 76,189 | https://github.com/openclaw/openclaw/issues/76189 | https://api.github.com/repos/openclaw/openclaw/issues/76189 | [Bug]: bncr gateway methods can become advertised-but-unknown again after SIGUSR1 in-process hot restart | ## Summary
On OpenClaw `v2026.4.26`, `bncr` is healthy after a full cold restart, but can still regress on `SIGUSR1` / `OPENCLAW_NO_RESPAWN` in-process hot restart: plugin logs show `bncr.connect` registered, yet live gateway WS requests later return `unknown method: bncr.connect`.
This looks like an uncovered varian... | closed | completed | false | 2 | [] | [] | 2026-05-02T17:46:21Z | 2026-05-03T08:01:16Z | 2026-05-03T08:01:16Z | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | xmoxmo | 112,841,961 | U_kgDOBrnU6Q | User | false |
openclaw/openclaw | 4,369,636,741 | I_kwDOQb6kR88AAAABBHNdhQ | 76,194 | https://github.com/openclaw/openclaw/issues/76194 | https://api.github.com/repos/openclaw/openclaw/issues/76194 | [Bug]: v2026.4.29 gateway fails on native Windows — bundled plugin runtime deps EPERM (MAX_PATH + npm/pnpm rename) | ## Bug type
Crash (process/app exits or hangs)
## Summary
After upgrading from v2026.4.27 to v2026.4.29 on native Windows 11 (Node 24.13.1, npm 11.8.0), the gateway fails to start with `Gateway failed to start: Error: Startup failed: required secrets are unavailable`. The real error is `EPERM: operation not permitte... | closed | completed | true | 1 | [] | [] | 2026-05-02T17:59:02Z | 2026-05-05T04:26:38Z | 2026-05-02T18:02:19Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | MilleniumGenAI | 66,120,693 | MDQ6VXNlcjY2MTIwNjkz | User | false |
openclaw/openclaw | 4,369,648,258 | I_kwDOQb6kR88AAAABBHOKgg | 76,196 | https://github.com/openclaw/openclaw/issues/76196 | https://api.github.com/repos/openclaw/openclaw/issues/76196 | sessions_spawn model override does not take effect — child always uses default model | ## Bug
`sessions_spawn` with `model` parameter reports `modelApplied: true` but the child session always uses the agent default model instead of the specified override.
## Steps to Reproduce
1. Set agent default model to e.g. `zai/glm-5-turbo`
2. Spawn a subagent with a different model:
```
sessions_spawn({
... | closed | completed | true | 3 | [] | [] | 2026-05-02T18:04:35Z | 2026-05-05T04:26:37Z | 2026-05-02T18:12:35Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | ClawsBot-OpenClaw | 272,544,191 | U_kgDOED6xvw | User | false |
openclaw/openclaw | 4,369,666,402 | I_kwDOQb6kR88AAAABBHPRYg | 76,198 | https://github.com/openclaw/openclaw/issues/76198 | https://api.github.com/repos/openclaw/openclaw/issues/76198 | Telegram channel fails to load: Cannot find module grammy (missing bundled dep) | ## Bug description
The Telegram channel fails to load on startup with the following error:
```
[channels] failed to load bundled channel telegram: Cannot find module 'grammy'
Require stack:
- /home/mike/.npm-global/lib/node_modules/openclaw/dist/extensions/telegram/send-t4GTTdHF.js
```
The extension file `dist/exten... | closed | completed | true | 1 | [] | [] | 2026-05-02T18:13:07Z | 2026-05-05T04:26:36Z | 2026-05-02T18:17:03Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | micheljdot-dot | 279,348,106 | U_kgDOEKaDig | User | false |
openclaw/openclaw | 4,369,667,540 | I_kwDOQb6kR88AAAABBHPV1A | 76,200 | https://github.com/openclaw/openclaw/issues/76200 | https://api.github.com/repos/openclaw/openclaw/issues/76200 | [Bug]: runtime-deps cache hit-rate ~13% on Pi 400 — staging blocks event-loop on every embedded-run (regression in 2026.4.29) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from 2026.4.27 to 2026.4.29 (with 2026.4.28 not present in CHANGELOG, possibly retracted), plugin-runtime-deps cache hit-rate dropped to ~13% on a Pi 400 deployment. Every embedded-run spawn re-installs npm-d... | closed | completed | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-02T18:13:39Z | 2026-05-05T04:26:35Z | 2026-05-02T18:17:59Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | mrschaal | 24,995,655 | MDQ6VXNlcjI0OTk1NjU1 | User | false |
openclaw/openclaw | 4,369,686,252 | I_kwDOQb6kR88AAAABBHQe7A | 76,202 | https://github.com/openclaw/openclaw/issues/76202 | https://api.github.com/repos/openclaw/openclaw/issues/76202 | [Bug]: channels.discord.commands.native=false still fetches slash commands on startup (GET /applications/.../commands times out) | ## Summary
Setting `channels.discord.commands.native: false` is supposed to disable native slash command deployment, but on startup the gateway still issues a `GET /applications/{appId}/commands` request. On a slow or congested network this request takes 30–40 seconds and blocks the Node.js event loop before it finall... | closed | completed | false | 2 | [] | [
"vincentkoc"
] | 2026-05-02T18:22:48Z | 2026-05-03T18:06:15Z | 2026-05-03T18:06:15Z | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | DBOYttt | 65,898,012 | MDQ6VXNlcjY1ODk4MDEy | User | false |
openclaw/openclaw | 4,369,693,147 | I_kwDOQb6kR88AAAABBHQ52w | 76,203 | https://github.com/openclaw/openclaw/issues/76203 | https://api.github.com/repos/openclaw/openclaw/issues/76203 | Performance regression: eager media/web provider discovery and uncached provider hook resolution add several seconds before streaming | ## Summary
A recent OpenClaw version appears to spend significant time before streaming begins, even when using a minimal tool profile. The delay is reproducible outside of Discord/message ingress and appears to occur inside OpenClaw's agent/runtime preparation path.
The main contributors observed were:
- eager cons... | closed | completed | true | 4 | [] | [] | 2026-05-02T18:26:02Z | 2026-05-05T04:25:41Z | 2026-05-02T23:39:37Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | flowforgelab | 206,646,681 | U_kgDODFEtmQ | User | false |
openclaw/openclaw | 4,369,700,535 | I_kwDOQb6kR88AAAABBHRWtw | 76,206 | https://github.com/openclaw/openclaw/issues/76206 | https://api.github.com/repos/openclaw/openclaw/issues/76206 | `openclaw status` does an unsafe `.trim()` on `runtimeModel` — crashes if any session entry has a non-string `model`/`modelProvider`/`providerOverride`/`modelOverride` | **Version:** confirmed in 2026.4.29 (active install at `/opt/homebrew/lib/node_modules/openclaw/dist/`); same pattern present in 2026.4.11 (npm-installed copy at `~/.nvm/.../openclaw/dist/`). First observed 2026-04-26 with a stale `claude-local` session entry that had a non-string `model`.
**Stack location:**
- Crash ... | closed | completed | false | 2 | [] | [
"vincentkoc"
] | 2026-05-02T18:29:42Z | 2026-05-03T19:30:57Z | 2026-05-03T19:30:47Z | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | intrix-dev | 269,304,916 | U_kgDOEA1EVA | User | false |
openclaw/openclaw | 4,369,700,473 | I_kwDOQb6kR88AAAABBHRWeQ | 76,205 | https://github.com/openclaw/openclaw/issues/76205 | https://api.github.com/repos/openclaw/openclaw/issues/76205 | Bundled runtime deps not declared in package.json — gateway auto-installs `chokidar`, `sqlite-vec` etc. on every boot | **Version:** observed in 2026.4.27 (sqlite-vec failed import) and 2026.4.29 (auto-install path appears to be the workaround; deps are now installed but still via the runtime-fallback mechanism, not via package.json declarations)
**Symptoms (2026.4.27):**
```
[memory] sqlite-vec unavailable: Cannot find package 'sqlite... | closed | completed | true | 1 | [] | [] | 2026-05-02T18:29:40Z | 2026-05-05T04:26:34Z | 2026-05-02T18:34:03Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | intrix-dev | 269,304,916 | U_kgDOEA1EVA | User | false |
openclaw/openclaw | 4,369,670,015 | I_kwDOQb6kR88AAAABBHPffw | 76,201 | https://github.com/openclaw/openclaw/issues/76201 | https://api.github.com/repos/openclaw/openclaw/issues/76201 | [Bug]: Plugin `before_tool_call` hook does not fire for native exec on 2026.4.29 (Anthropic harness) | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
A plugin that registers a `before_tool_call` hook never sees that handler invoked when a native tool (e.g. `exec`) runs through the Anthropic harness on `2026.4.29` (commit `a448042`). The plugin's `register()`... | open | null | false | 1 | [] | [] | 2026-05-02T18:14:51Z | 2026-05-03T05:01:07Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | juno02139 | 269,345,962 | U_kgDOEA3kqg | User | false |
openclaw/openclaw | 4,369,715,853 | I_kwDOQb6kR88AAAABBHSSjQ | 76,209 | https://github.com/openclaw/openclaw/issues/76209 | https://api.github.com/repos/openclaw/openclaw/issues/76209 | [Bug]: | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
<body>
<!--StartFragment--><h2 class="text-text-100 mt-3 -mb-1 text-[1.375rem] font-bold">Cache <code class="bg-text-200/5 border border-0.5 border-border-300 text-danger-000 whitespace-pre-wrap rounded-[0.4re... | open | null | false | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-05-02T18:36:49Z | 2026-05-03T05:01:12Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | shenhonglong456-ai | 271,716,315 | U_kgDOEDIP2w | User | false |
openclaw/openclaw | 4,369,711,675 | I_kwDOQb6kR88AAAABBHSCOw | 76,208 | https://github.com/openclaw/openclaw/issues/76208 | https://api.github.com/repos/openclaw/openclaw/issues/76208 | agents.create triggers gateway restart, breaking in-flight calls (~1.5–2s port unavailability) | ## Summary
When a backend client calls `agents.create` over WebSocket, the gateway writes the new agent's auth token to `<state>/devices/paired.json`. The config-watcher detects this change and concludes a gateway restart is needed; SIGUSR1 fires immediately, the gateway shuts down cleanly, then re-binds — closing por... | closed | completed | true | 1 | [] | [] | 2026-05-02T18:34:42Z | 2026-05-05T04:26:34Z | 2026-05-02T18:50:47Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | jldmentagent | 274,719,439 | U_kgDOEF_izw | User | false |
openclaw/openclaw | 4,369,737,446 | I_kwDOQb6kR88AAAABBHTm5g | 76,212 | https://github.com/openclaw/openclaw/issues/76212 | https://api.github.com/repos/openclaw/openclaw/issues/76212 | [Bug]: 2026.4.29 gateway instability: Active Memory timeouts, embedded-run prep latency, event-loop pressure | ### Bug type
Regression (worked better before, now unstable/slow)
### Summary
After upgrading to OpenClaw `2026.4.29`, the gateway became noticeably unstable and slow. The main symptoms are high per-turn latency, Active Memory timeouts, repeated embedded-run startup/prep overhead, event-loop delay warnings, and slow... | closed | not_planned | true | 1 | [] | [] | 2026-05-02T18:47:14Z | 2026-05-05T04:26:33Z | 2026-05-02T18:51:04Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | harleymdsavage | 258,685,029 | U_kgDOD2s4ZQ | User | false |
openclaw/openclaw | 4,369,736,723 | I_kwDOQb6kR88AAAABBHTkEw | 76,211 | https://github.com/openclaw/openclaw/issues/76211 | https://api.github.com/repos/openclaw/openclaw/issues/76211 | [Bug] Codex OAuth token fails in isolated cron/heartbeat sessions — atob URL-safe base64 issue | ## Bug Description
OpenAI Codex (codex/gpt-5.5) fails with `Failed to extract accountId from token` when used in isolated cron or heartbeat sessions, despite working correctly in the main interactive session.
## Environment
- OpenClaw: 2026.4.29 (a448042)
- Platform: macOS (Darwin 25.2.0, Apple Silicon)
- Node.js: v... | closed | completed | true | 1 | [] | [] | 2026-05-02T18:47:00Z | 2026-05-05T04:26:32Z | 2026-05-02T18:52:52Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | ryancush | 964,630 | MDQ6VXNlcjk2NDYzMA== | User | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.