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 | 4,208,154,291 | I_kwDOQb6kR87601az | 61,447 | https://github.com/openclaw/openclaw/issues/61447 | https://api.github.com/repos/openclaw/openclaw/issues/61447 | feat(hooks): include compaction summary text in after_compaction hook payload | ## Problem
The `after_compaction` hook event currently provides basic metadata (session key, message counts) but does not include the actual compaction summary text that was generated.
Plugins that want to audit compaction quality, score signal preservation, or extract key facts from the summary have no way to access... | open | null | false | 1 | [] | [] | 2026-04-05T17:23:54Z | 2026-05-02T09:00:08Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | 5p00kyy | 20,588,316 | MDQ6VXNlcjIwNTg4MzE2 | User | false |
openclaw/openclaw | 4,209,456,180 | I_kwDOQb6kR8765zQ0 | 61,673 | https://github.com/openclaw/openclaw/issues/61673 | https://api.github.com/repos/openclaw/openclaw/issues/61673 | [Feature]: Pre-container-creation hook for dynamic sandbox bind mounts | ### Summary
Allow a host-side hook (script or callback) to run before a sandbox container is created, enabling dynamic resolution of `sandbox.docker.binds` based on session context (channel, user, metadata).
### Problem to solve
In multi-user gateway deployments with differentiated file access (governance levels, co... | open | null | false | 1 | [] | [] | 2026-04-06T04:08:14Z | 2026-05-02T09:22:21Z | null | NONE | null | 20260502T120029Z | 2026-05-02T12:00:29Z | rodja | 131,391 | MDQ6VXNlcjEzMTM5MQ== | User | false |
openclaw/openclaw | 4,208,585,765 | I_kwDOQb6kR8762ewl | 61,520 | https://github.com/openclaw/openclaw/issues/61520 | https://api.github.com/repos/openclaw/openclaw/issues/61520 | [Feature]: Config IO silently swallows errors that may indicate serious problems | ### Summary
The config IO system (`src/config/io.ts`) contains multiple catch blocks that silently swallow errors without proper logging or error recovery. This makes debugging configuration issues extremely difficult, as failures in config writes, health state updates, and audit logging are completely invisible to op... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-05T21:11:29Z | 2026-05-02T06:09:35Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | soumojit-D48 | 207,209,292 | U_kgDODFnDTA | User | false |
openclaw/openclaw | 4,217,927,561 | I_kwDOQb6kR877aHeJ | 62,513 | https://github.com/openclaw/openclaw/issues/62513 | https://api.github.com/repos/openclaw/openclaw/issues/62513 | [Feature]: this.hasIndexedContent() seems to be called too frequently, causing performance issus | ### Summary
```
async search(
query: string,
opts?: {
maxResults?: number;
minScore?: number;
sessionKey?: string;
},
): Promise<MemorySearchResult[]> {
const preflight = resolveMemorySearchPreflight({
query,
hasIndexedContent: this.hasIndexedContent(),
});
if (!... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-07T13:17:56Z | 2026-05-02T09:00:13Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | lixianmin | 26,290,856 | MDQ6VXNlcjI2MjkwODU2 | User | false |
openclaw/openclaw | 4,209,134,758 | I_kwDOQb6kR8764kym | 61,621 | https://github.com/openclaw/openclaw/issues/61621 | https://api.github.com/repos/openclaw/openclaw/issues/61621 | 飞书消息引用时显示内部元数据标签（message_id、sender_id 等） | ## 问题描述
在飞书聊天中，当用户引用/回复消息时，OpenClaw 的内部元数据标签（如 `[message_id: xxx]`、`ou_xxx` 等）会显示在用户可见的消息内容中。
这些标签原本是给 AI agent 的上下文信息，但在某些场景下会意外暴露给用户，影响阅读体验。
## 期望行为
1. 这些内部元数据标签应该对用户不可见
2. 或者提供一个配置选项让用户可以选择是否显示这些标签
## 当前行为
用户在飞书中看到类似这样的内容：
```
[message_id: om_x100b520521b350a0c110612894d405a]
ou_ad5e4b2549c6a81025a166986a6b914... | open | null | false | 1 | [] | [] | 2026-04-06T02:20:59Z | 2026-05-02T07:02:04Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | xiaojiayao | 42,060,404 | MDQ6VXNlcjQyMDYwNDA0 | User | false |
openclaw/openclaw | 4,200,767,555 | I_kwDOQb6kR876YqBD | 60,335 | https://github.com/openclaw/openclaw/issues/60335 | https://api.github.com/repos/openclaw/openclaw/issues/60335 | Slack: thread replies re-attach parent message media on every reply | ## Bug: Slack thread replies re-attach parent message media
When a Slack message containing an image starts a thread, every subsequent reply in that thread re-delivers the original image to the model — even when the reply contains no media of its own.
### Impact
- **Token waste:** Every thread reply incurs vision mo... | open | null | false | 2 | [] | [] | 2026-04-03T14:20:32Z | 2026-05-02T07:01:59Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | rachelbotley | 268,250,971 | U_kgDOD_0vWw | User | false |
openclaw/openclaw | 4,212,957,633 | I_kwDOQb6kR877HKHB | 62,029 | https://github.com/openclaw/openclaw/issues/62029 | https://api.github.com/repos/openclaw/openclaw/issues/62029 | [Feature]: Dreaming: add wildcard candidates to deep phase ranking | ### Summary
Add random sub-threshold wildcard candidates to dreaming deep phase ranking so that unexpected memories can occasionally surface for promotion
### Problem to solve
The current dreaming system only promotes memories that score well on a convergence-oriented model — high recall frequency, strong relevance,... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-06T17:05:59Z | 2026-05-02T06:31:10Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | recozers | 115,054,941 | U_kgDOBtuZXQ | User | false |
openclaw/openclaw | 4,207,620,444 | I_kwDOQb6kR876yzFc | 61,368 | https://github.com/openclaw/openclaw/issues/61368 | https://api.github.com/repos/openclaw/openclaw/issues/61368 | [Feature]: Full Workspace File Management API for Remote/Container Deployments | ### Summary
As OpenClaw adoption grows in containerized and cloud-native environments, users increasingly need a way to manage agent workspace files without direct filesystem access. The current `agents.files.*` API restricts file operations to a predefined whitelist of bootstrap files (AGENTS.md, MEMORY.md, etc.), wh... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-05T13:59:52Z | 2026-05-02T11:08:32Z | null | NONE | null | 20260502T120029Z | 2026-05-02T12:00:29Z | WilShi | 47,671,779 | MDQ6VXNlcjQ3NjcxNzc5 | User | false |
openclaw/openclaw | 4,211,268,769 | I_kwDOQb6kR877Atyh | 61,831 | https://github.com/openclaw/openclaw/issues/61831 | https://api.github.com/repos/openclaw/openclaw/issues/61831 | [Bug] Every agent shares one Chrome instance — no browser isolation | ## Bug Description
All OpenClaw agents share a single Chrome instance when using the browser tool, with no isolation between agents. This creates two related problems:
### 1. No per-agent browser process
When multiple agents (e.g., Silas, Hector, Ben, Tony) use the browser tool simultaneously, they all connect to the... | open | null | false | 1 | [] | [] | 2026-04-06T11:10:25Z | 2026-05-02T09:22:22Z | null | NONE | null | 20260502T120029Z | 2026-05-02T12:00:29Z | ttomiczek | 265,786,127 | U_kgDOD9eTDw | User | false |
openclaw/openclaw | 4,189,435,536 | I_kwDOQb6kR875tbaQ | 59,225 | https://github.com/openclaw/openclaw/issues/59225 | https://api.github.com/repos/openclaw/openclaw/issues/59225 | Feature: Per-agent tool schema parameter filtering (slim tool definitions for subagents) | ## Problem
When non-main agents (e.g., a mechanical dispatcher agent) call `sessions_spawn`, every model fills in ALL optional parameters from the tool schema — `sandbox`, `streamTo`, `attachAs`, `cleanup`, `attachments`, `runtime`, `label`, `thread`, `model`, `thinking`, `cwd` — even when explicitly instructed not to... | open | null | false | 1 | [] | [] | 2026-04-01T19:40:33Z | 2026-05-01T00:44:58Z | null | NONE | null | 20260501T113057Z | 2026-05-01T11:30:57Z | JanPlessow | 202,702,070 | U_kgDODBT89g | User | false |
openclaw/openclaw | 4,208,363,178 | I_kwDOQb6kR8761oaq | 61,484 | https://github.com/openclaw/openclaw/issues/61484 | https://api.github.com/repos/openclaw/openclaw/issues/61484 | [Feature Request] Support image/media files between agents via sessions_send | ## Feature Request: Support image/media files between agents via sessions_send
### Background
Currently `sessions_send` only supports plain text, making it impossible to send images directly between agents (e.g., main agent <-> reviewer agent). This forces teams to use third-party image hosts as a workaround.
### Req... | open | null | false | 1 | [] | [] | 2026-04-05T19:08:38Z | 2026-05-02T07:49:53Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | ryswork1993 | 44,529,247 | MDQ6VXNlcjQ0NTI5MjQ3 | User | false |
openclaw/openclaw | 4,208,044,149 | I_kwDOQb6kR8760ah1 | 61,430 | https://github.com/openclaw/openclaw/issues/61430 | https://api.github.com/repos/openclaw/openclaw/issues/61430 | feat(exec): add description field for intent-based hook approval | ## Problem
Currently the exec tool only carries a `command` string. Hook systems like permission-enforcers must use pattern matching on the command text to decide whether to allow or block an operation. This is fragile and doesn't capture the user's intent.
## Proposed Solution
Add a `description` (or `intent`) fiel... | open | null | false | 1 | [] | [] | 2026-04-05T16:33:43Z | 2026-05-02T09:22:18Z | null | NONE | null | 20260502T120029Z | 2026-05-02T12:00:29Z | chamichen-design | 270,144,372 | U_kgDOEBoTdA | User | false |
openclaw/openclaw | 4,217,540,027 | I_kwDOQb6kR877Yo27 | 62,464 | https://github.com/openclaw/openclaw/issues/62464 | https://api.github.com/repos/openclaw/openclaw/issues/62464 | [Bug] MiniMax API timeout causes complete request failure with no retry | ## Bug Description
MiniMax API (both M2.7 and M2.7-highspeed) is timing out during requests, causing complete failure with no retry. The fallback chain exhausts both models and then fails.
## Frequency
4+ times in 24h period (2026-04-06 to 2026-04-07)
## Log Evidence
```
[agent/embedded] embedded run failover decisio... | open | null | false | 1 | [] | [] | 2026-04-07T12:08:42Z | 2026-05-02T07:02:13Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | jaydenlichong29-create | 268,866,858 | U_kgDOEAaVKg | User | false |
openclaw/openclaw | 4,213,474,576 | I_kwDOQb6kR877JIUQ | 62,077 | https://github.com/openclaw/openclaw/issues/62077 | https://api.github.com/repos/openclaw/openclaw/issues/62077 | Feature: Minimal inbound metadata mode for DM (direct) chats to reduce context bloat and hallucinations | ## Problem
In Telegram DM (1:1) chats, `buildInboundUserContextPrefix` injects **full metadata blocks** with every single user message:
```
Conversation info (untrusted metadata):
{"message_id": "697", "sender_id": "292593739", "sender": "Vit Gruver", "timestamp": "Mon 2026-04-06 16:55 UTC"}
Sender (untrusted metada... | open | null | false | 1 | [] | [] | 2026-04-06T18:51:04Z | 2026-05-02T05:56:24Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | VitGruver | 85,167,831 | MDQ6VXNlcjg1MTY3ODMx | User | false |
openclaw/openclaw | 4,208,027,555 | I_kwDOQb6kR8760Wej | 61,426 | https://github.com/openclaw/openclaw/issues/61426 | https://api.github.com/repos/openclaw/openclaw/issues/61426 | [Bug]: sessionTarget: isolated not honored — cron job messages accumulate in main session | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Cron jobs configured with sessionTarget: isolated are not running in isolated sessions. Instead, their messages accumulate silently in the main conversation session (conversation_id 6 in lcm.db), causing context overflow and... | closed | completed | true | 2 | [
"bug",
"regression"
] | [] | 2026-04-05T16:24:47Z | 2026-05-05T04:28:29Z | 2026-05-02T06:41:59Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Lihannon | 268,415,597 | U_kgDOD_-ybQ | User | false |
openclaw/openclaw | 4,215,116,116 | I_kwDOQb6kR877PZFU | 62,261 | https://github.com/openclaw/openclaw/issues/62261 | https://api.github.com/repos/openclaw/openclaw/issues/62261 | [Bug]: Dev channel fails to build (preflight lint fails) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
Yes
### Summary
Stable release (last: 2026.4.5) was working fine.
Tried to switch to dev update channel, `openclaw update --channel dev`
Got a few "preflight lint" errors (different commit hashes) that look like:
```
✗ preflight lint (5... | open | null | false | 3 | [
"bug",
"regression"
] | [] | 2026-04-07T02:48:14Z | 2026-05-02T07:16:20Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | tturnerdev | 1,414,401 | MDQ6VXNlcjE0MTQ0MDE= | User | false |
openclaw/openclaw | 4,212,622,763 | I_kwDOQb6kR877F4Wr | 61,986 | https://github.com/openclaw/openclaw/issues/61986 | https://api.github.com/repos/openclaw/openclaw/issues/61986 | [Bug]: Android companion notification events trigger individual agent runs instead of batching to heartbeat interval | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Android companion notification events each trigger an independent agent run instead of being batched to the heartbeat interval, causing unbounded API consumption.
### Steps to reproduce
Configure a heartbeat ... | open | null | false | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-04-06T15:55:40Z | 2026-05-02T07:49:56Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | mogglemoss | 10,966,195 | MDQ6VXNlcjEwOTY2MTk1 | User | false |
openclaw/openclaw | 4,357,973,952 | I_kwDOQb6kR88AAAABA8FnwA | 75,068 | https://github.com/openclaw/openclaw/issues/75068 | https://api.github.com/repos/openclaw/openclaw/issues/75068 | Bug: Brave web_search falsely rejects freshness, date_after/date_before, and ui_lang in llm-context mode | # Bug Report: Brave `web_search` falsely rejects `freshness`, `date_after/date_before`, and `ui_lang` in `llm-context` mode
## Summary
OpenClaw's `web_search` tool with Brave provider in `llm-context` mode artificially blocks parameters that the **Brave LLM Context API endpoint fully supports**. The validation logic ... | closed | not_planned | true | 1 | [] | [] | 2026-04-30T12:13:30Z | 2026-05-04T04:34:42Z | 2026-05-02T01:48:29Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | vokasug | 280,127,440 | U_kgDOELJn0A | User | false |
openclaw/openclaw | 4,357,975,235 | I_kwDOQb6kR88AAAABA8Fsww | 75,070 | https://github.com/openclaw/openclaw/issues/75070 | https://api.github.com/repos/openclaw/openclaw/issues/75070 | [Bug]: Model hallucinates fabricated data on session reset | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
On session reset, the model receives an empty or bootstrap prompt but
instead of generating a simple welcome message, it fabricates
realistic-looking data (calendar events, JSON structures) and presents
i... | open | null | false | 1 | [
"bug"
] | [] | 2026-04-30T12:13:42Z | 2026-05-03T06:08:12Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | Lkrambar | 158,211,079 | U_kgDOCW4cBw | User | false |
openclaw/openclaw | 4,357,904,226 | I_kwDOQb6kR88AAAABA8BXYg | 75,060 | https://github.com/openclaw/openclaw/issues/75060 | https://api.github.com/repos/openclaw/openclaw/issues/75060 | [Bug]: execution "direct" config silently reverted when gateway.bind="lan" | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
### Environment
- OpenClaw version: 2026.4.27
- Platform: Linux (ESXi 8.0, Debian 12)
- Gateway bind: "lan"
- Config file: ~/.openclaw/openclaw.json
### Problem
When setting `execution: "direct"` in `agents.defaults` (or u... | closed | completed | true | 2 | [
"bug",
"regression"
] | [] | 2026-04-30T12:02:25Z | 2026-05-03T04:35:00Z | 2026-04-30T21:14:35Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | onridges | 17,865,300 | MDQ6VXNlcjE3ODY1MzAw | User | false |
openclaw/openclaw | 4,215,331,941 | I_kwDOQb6kR877QNxl | 62,284 | https://github.com/openclaw/openclaw/issues/62284 | https://api.github.com/repos/openclaw/openclaw/issues/62284 | feat: time-bounded exec approvals (approve for N minutes/hours) | ## Summary
Add a time-bounded approval mode so users can approve a command pattern for a specified duration (e.g. 30 minutes, 1 hour) without having to approve every single invocation or permanently allow it.
## Motivation
The current approval modes are:
- `once` — approve this command once, prompt again next time
-... | open | null | false | 1 | [] | [] | 2026-04-07T04:06:52Z | 2026-05-02T06:31:17Z | null | CONTRIBUTOR | null | 20260502T090030Z | 2026-05-02T09:00:30Z | scottgl9 | 10,623,288 | MDQ6VXNlcjEwNjIzMjg4 | User | false |
openclaw/openclaw | 4,216,925,632 | I_kwDOQb6kR877WS3A | 62,420 | https://github.com/openclaw/openclaw/issues/62420 | https://api.github.com/repos/openclaw/openclaw/issues/62420 | Bug: in-turn auto-compaction can bypass memoryFlush because memoryFlush only runs pre-turn | ## Summary
When `agents.defaults.compaction.memoryFlush.enabled=true`, OpenClaw can still perform an **in-turn auto-compaction** without writing any memory files.
The root cause appears to be that `runMemoryFlushIfNeeded()` only runs **before** the turn starts, while some auto-compactions happen **during** the turn. ... | open | null | false | 1 | [] | [] | 2026-04-07T10:03:23Z | 2026-05-02T07:02:12Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | realaboo | 756,485 | MDQ6VXNlcjc1NjQ4NQ== | User | false |
openclaw/openclaw | 4,211,290,127 | I_kwDOQb6kR877AzAP | 61,834 | https://github.com/openclaw/openclaw/issues/61834 | https://api.github.com/repos/openclaw/openclaw/issues/61834 | [Feature]: expose QMD no-rerank for memory.qmd query mode | ## Summary
Expose QMD 2.1.0's no-rerank capability from OpenClaw's QMD memory backend so users can run `searchMode: "query"` without the reranking step.
## Problem to solve
OpenClaw already supports:
- `memory.backend = "qmd"`
- `memory.qmd.searchMode = "query" | "search" | "vsearch"`
But when `searchMode: "query"... | open | null | false | 1 | [] | [] | 2026-04-06T11:14:53Z | 2026-05-02T06:31:08Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | kouka-t0yohei | 261,699,157 | U_kgDOD5k2VQ | User | false |
openclaw/openclaw | 4,210,033,416 | I_kwDOQb6kR8768AMI | 61,723 | https://github.com/openclaw/openclaw/issues/61723 | https://api.github.com/repos/openclaw/openclaw/issues/61723 | [Bug]: TUI process starts but UI does not render after gateway restart or kill -9 | ### Bug type
Crash (process/app exits or hangs)
### Beta release blocker
No
### Summary
After restarting the gateway via `systemctl --user restart openclaw-gateway` or killing the TUI process with `kill -9`, subsequent `openclaw tui` invocations start the process and consume ~900MB of memory but never render the i... | closed | not_planned | true | 2 | [] | [] | 2026-04-06T06:35:33Z | 2026-05-03T04:34:36Z | 2026-05-01T00:08:31Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | mslugyfw | 40,541,429 | MDQ6VXNlcjQwNTQxNDI5 | User | false |
openclaw/openclaw | 4,217,896,212 | I_kwDOQb6kR877Z_0U | 62,512 | https://github.com/openclaw/openclaw/issues/62512 | https://api.github.com/repos/openclaw/openclaw/issues/62512 | AI-powered in-app help that guides users by walking them through the UI | ## The idea
Help in most apps is a dead end: a search box, a docs site in another tab, a tooltip pointing at the wrong thing. People don't want to *read* how to do something — they want someone to *show* them.
This issue proposes a new kind of help for OpenClaw: an AI guide that lives inside the app, listens to what ... | open | null | false | 1 | [] | [] | 2026-04-07T13:12:02Z | 2026-05-02T06:31:27Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | LeVarez | 76,524,987 | MDQ6VXNlcjc2NTI0OTg3 | User | false |
openclaw/openclaw | 4,211,889,478 | I_kwDOQb6kR877DFVG | 61,895 | https://github.com/openclaw/openclaw/issues/61895 | https://api.github.com/repos/openclaw/openclaw/issues/61895 | [Bug]: Persistent ACP sessions accumulate orphan claude-agent-acp processes after ensureSession repair — progressive memory leak | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Persistent ACP sessions (bound to Telegram topics/DMs via `mode: "persistent"`) accumulate orphan `claude-agent-acp` and `plugin-tools-serve.js` processes over time, leading to progressive memory exhaustion and... | open | null | false | 2 | [] | [] | 2026-04-06T13:26:18Z | 2026-05-02T06:09:41Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | richardmqq | 14,178,267 | MDQ6VXNlcjE0MTc4MjY3 | User | false |
openclaw/openclaw | 4,214,459,932 | I_kwDOQb6kR877M44c | 62,167 | https://github.com/openclaw/openclaw/issues/62167 | https://api.github.com/repos/openclaw/openclaw/issues/62167 | [Feature]: token usage widged in chat UI | ### Summary
Display token usage/remaining quota prominently in the OpenClaw Control UI chat interface.
Users would like:
A persistent token counter widget at the top of the chat
Per-message token usage shown below each assistant response
Providers to support: OpenRouter (credits API), LLMAPI (unlimited), Ollama (loc... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-06T22:50:43Z | 2026-05-02T06:31:15Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | antritax23 | 241,370,247 | U_kgDODmMEhw | User | false |
openclaw/openclaw | 4,213,875,702 | I_kwDOQb6kR877KqP2 | 62,109 | https://github.com/openclaw/openclaw/issues/62109 | https://api.github.com/repos/openclaw/openclaw/issues/62109 | Interactive runs fail with auth-style 403 when custom OpenAI-compatible provider baseUrl uses Unicode/IDN or punycode hostname, but ASCII hostname/IP works | ### Summary
OpenClaw interactive runs (Web UI and local TUI / embedded path) fail with:
- `403 Your request was blocked.`
- `failoverReason: "auth"`
- `profileFailureReason: "auth"`
when a custom OpenAI-compatible provider uses either:
- a Unicode/IDN hostname in `baseUrl`
- or the punycode form of that same hostna... | open | null | false | 1 | [] | [] | 2026-04-06T20:20:27Z | 2026-05-02T05:56:27Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | nboody | 21,004,053 | MDQ6VXNlcjIxMDA0MDUz | User | false |
openclaw/openclaw | 4,213,202,294 | I_kwDOQb6kR877IF12 | 62,056 | https://github.com/openclaw/openclaw/issues/62056 | https://api.github.com/repos/openclaw/openclaw/issues/62056 | Feature: Expose Feishu bot added/removed events as hook events | ## Summary
When a Feishu bot is added to or removed from a group chat, OpenClaw already receives and processes the `im.chat.member.bot.added_v1` and `im.chat.member.bot.deleted_v1` events from the Feishu Event Subscription API. However, these events are only logged internally and not exposed to the hook system.
## Cu... | open | null | false | 1 | [] | [] | 2026-04-06T17:59:27Z | 2026-05-02T05:56:23Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | easydigi | 35,327,287 | MDQ6VXNlcjM1MzI3Mjg3 | User | false |
openclaw/openclaw | 4,209,984,971 | I_kwDOQb6kR87670XL | 61,716 | https://github.com/openclaw/openclaw/issues/61716 | https://api.github.com/repos/openclaw/openclaw/issues/61716 | [Feature]: Add model parameter prompts (context window, max_tokens, modalities) during OpenAI-compatible provider onboarding CLI flow | ### Summary
Add interactive model parameter prompts (context window, max output tokens, modalities) during OpenAI-compatible provider onboarding.
### Problem to solve
When running `openclaw onboard` and selecting an OpenAI-compatible provider, users can pick a model but cannot configure model-specific parameters. Se... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-06T06:22:46Z | 2026-05-02T05:56:22Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | Andy-Xie-1145 | 236,527,195 | U_kgDODhkeWw | User | false |
openclaw/openclaw | 4,215,683,595 | I_kwDOQb6kR877RjoL | 62,317 | https://github.com/openclaw/openclaw/issues/62317 | https://api.github.com/repos/openclaw/openclaw/issues/62317 | /model status shows phantom provider extracted from OpenRouter model ID prefix | ## Description
When configuring an OpenRouter model whose ID contains a known provider prefix (e.g. `google/gemini-3-flash-preview`), `/model status` displays a phantom `[google]` provider entry alongside the intended `[openrouter]` entry.
## Steps to reproduce
1. Configure `models.json` with only an `openrouter` pr... | open | null | false | 1 | [] | [] | 2026-04-07T05:38:14Z | 2026-05-02T06:31:20Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | firesh | 7,334,304 | MDQ6VXNlcjczMzQzMDQ= | User | false |
openclaw/openclaw | 4,217,555,350 | I_kwDOQb6kR877YsmW | 62,466 | https://github.com/openclaw/openclaw/issues/62466 | https://api.github.com/repos/openclaw/openclaw/issues/62466 | [Bug] read tool allows offset beyond file length causing crash | ## Bug Description
The read tool accepts an offset parameter that exceeds the file's actual line count, causing the tool to fail with 'Offset X is beyond end of file'.
## Log Evidence
```
[tools] read failed: Offset 1830 is beyond end of file (1829 lines total) raw_params={"offset":1830,"limit":20,"file_path":"/Users/... | open | null | false | 1 | [] | [] | 2026-04-07T12:11:57Z | 2026-05-02T06:31:25Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | jaydenlichong29-create | 268,866,858 | U_kgDOEAaVKg | User | false |
openclaw/openclaw | 4,211,381,416 | I_kwDOQb6kR877BJSo | 61,840 | https://github.com/openclaw/openclaw/issues/61840 | https://api.github.com/repos/openclaw/openclaw/issues/61840 | [Bug]: | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
问题：sendMediaFeishu 调用 uploadFileFeishu 时没有传递 duration 参数
影响：飞书语音消息时长显示为 0 秒
▎ sendMediaFeishu 发送飞书语音消息时未传递 duration 参数
... | closed | not_planned | true | 1 | [
"bug",
"regression"
] | [] | 2026-04-06T11:33:27Z | 2026-05-03T04:37:36Z | 2026-04-30T12:49:26Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | fxz26284407 | 34,901,544 | MDQ6VXNlcjM0OTAxNTQ0 | User | false |
openclaw/openclaw | 4,212,856,112 | I_kwDOQb6kR877GxUw | 62,022 | https://github.com/openclaw/openclaw/issues/62022 | https://api.github.com/repos/openclaw/openclaw/issues/62022 | Codex profile rotation skips remaining profiles and jumps to fallback model chain on rate limit | ## Bug Description
When a Codex profile in the rotation order returns a rate limit error, OpenClaw skips the remaining profiles in the rotation and immediately escalates to the **next model in the fallback chain** (e.g., from `gpt-5.3-codex` to `*next_fallback_model*`), rather than continuing to try the remaining prof... | open | null | false | 2 | [] | [] | 2026-04-06T16:44:53Z | 2026-05-02T07:16:19Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | DeadlySilent | 3,342,508 | MDQ6VXNlcjMzNDI1MDg= | User | false |
openclaw/openclaw | 4,357,992,775 | I_kwDOQb6kR88AAAABA8GxRw | 75,072 | https://github.com/openclaw/openclaw/issues/75072 | https://api.github.com/repos/openclaw/openclaw/issues/75072 | Feature request: PII sanitization skill for privacy compliance before LLM calls | ### Summary
OpenClaw processes user messages from WhatsApp, Telegram
and Discord that frequently contain PII. TrustBoost is
a ready-made sanitization skill that redacts sensitive
data before it reaches LLMs.
### Problem to solve
User messages sent through OpenClaw channels contain
sensitive PII — emails, phone n... | closed | not_planned | true | 1 | [
"enhancement"
] | [] | 2026-04-30T12:16:35Z | 2026-05-03T04:37:36Z | 2026-04-30T12:49:31Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | teodorofodocrispin-cmyk | 271,404,169 | U_kgDOEC1MiQ | User | false |
openclaw/openclaw | 4,214,218,147 | I_kwDOQb6kR877L92j | 62,145 | https://github.com/openclaw/openclaw/issues/62145 | https://api.github.com/repos/openclaw/openclaw/issues/62145 | [Feature]: Rethink login-gate | ### Summary
Often when the agent is doing something with the platform, and takes down the gateway, it does so while you are reading the messages. Or need to copy a command.
### Problem to solve
The control UI instantly disconnects, blocking access to any of the prior messages.
### Proposed solution
It would be ni... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-06T21:40:56Z | 2026-05-02T10:36:40Z | null | CONTRIBUTOR | null | 20260502T120029Z | 2026-05-02T12:00:29Z | bernesto | 1,020,507 | MDQ6VXNlcjEwMjA1MDc= | User | false |
openclaw/openclaw | 4,215,389,601 | I_kwDOQb6kR877Qb2h | 62,294 | https://github.com/openclaw/openclaw/issues/62294 | https://api.github.com/repos/openclaw/openclaw/issues/62294 | Heartbeat: non-interval wake reasons bypass interval enforcement, causing bursts and silent gaps | ## Summary
Non-interval heartbeat wake reasons (`exec-event`, `wake`, `hook`, `cron:*`, `background-task`, `notifications-event`) bypass the per-agent interval gate entirely, causing two related problems:
1. **Burst runs** — multiple heartbeats fire in rapid succession when activity resumes after idle periods
2. **Si... | open | null | false | 4 | [] | [] | 2026-04-07T04:23:51Z | 2026-05-04T10:38:31Z | null | NONE | null | 20260504T180039Z | 2026-05-04T18:00:39Z | bryanpearson | 926,346 | MDQ6VXNlcjkyNjM0Ng== | User | false |
openclaw/openclaw | 4,209,684,106 | I_kwDOQb6kR8766q6K | 61,694 | https://github.com/openclaw/openclaw/issues/61694 | https://api.github.com/repos/openclaw/openclaw/issues/61694 | [Feature]: Add opt-out for external CLI credential sync into auth profiles | ## Summary
Add a supported global opt-out for external CLI credential sync into OpenClaw auth profiles.
## Problem to solve
OpenClaw currently syncs some external CLI credentials into the auth-profile store at runtime, including Codex CLI credentials. That is convenient in some setups, but it causes the wrong behavi... | closed | duplicate | true | 2 | [] | [] | 2026-04-06T05:08:44Z | 2026-05-03T04:35:28Z | 2026-04-30T18:53:56Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | eilonkr | 8,035,202 | MDQ6VXNlcjgwMzUyMDI= | User | false |
openclaw/openclaw | 4,209,740,887 | I_kwDOQb6kR87664xX | 61,697 | https://github.com/openclaw/openclaw/issues/61697 | https://api.github.com/repos/openclaw/openclaw/issues/61697 | [Bug]: gog gmail watch serve + Pub/Sub: active watch but no /hooks/gmail deliveries on OpenClaw 2026.4.5 | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Gmail watch via gog gmail watch serve and Pub/Sub appears active and healthy, but no Gmail webhook requests ever reach OpenClaw’s /hooks/gmail endpoint.
### Steps to reproduce
1. Install OpenClaw 2026.4.5 on Windows 10 wit... | open | null | false | 2 | [
"bug",
"regression"
] | [] | 2026-04-06T05:22:44Z | 2026-04-30T12:49:54Z | null | NONE | null | 20260430T150026Z | 2026-04-30T15:00:26Z | buttheadus | 85,201,953 | MDQ6VXNlcjg1MjAxOTUz | User | false |
openclaw/openclaw | 4,207,999,003 | I_kwDOQb6kR8760Pgb | 61,418 | https://github.com/openclaw/openclaw/issues/61418 | https://api.github.com/repos/openclaw/openclaw/issues/61418 | [Feature Request]: tUI /models command should filter by auth status | ## Problem
When running the `/models` command in the OpenClaw TUI, the model list shows **hundreds of models** that have no API key configured. Only 1-2 models (e.g. minimax) have auth configured.
In contrast, running `openclaw models list` in the terminal correctly shows only models with configured auth credentials.... | open | null | false | 1 | [] | [] | 2026-04-05T16:11:47Z | 2026-05-02T06:09:34Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | zamanewby | 97,544 | MDQ6VXNlcjk3NTQ0 | User | false |
openclaw/openclaw | 4,210,327,185 | I_kwDOQb6kR8769H6R | 61,740 | https://github.com/openclaw/openclaw/issues/61740 | https://api.github.com/repos/openclaw/openclaw/issues/61740 | Feature Request: Runtime Skill Discovery for Better Agent-Skill Matching | # Proposal: Runtime Skill Discovery for OpenClaw
**Date:** 2026-04-06
**From:** OpenClaw User (local-manager agent workspace)
**To:** OpenClaw Team
---
## Problem Statement
OpenClaw's current skill system is **static at session startup**:
1. Skills are loaded once when a session starts
2. Agent sees only `<availab... | open | null | false | 1 | [] | [] | 2026-04-06T07:40:53Z | 2026-05-02T06:09:37Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | eltonwf-del | 266,263,235 | U_kgDOD97aww | User | false |
openclaw/openclaw | 4,216,358,255 | I_kwDOQb6kR877UIVv | 62,387 | https://github.com/openclaw/openclaw/issues/62387 | https://api.github.com/repos/openclaw/openclaw/issues/62387 | fix(channels): most channels missing namedAccountPromotionKeys — multi-account promotion strips shared defaults | ## Summary
When a channel config has named accounts but no `default`, the doctor/`channels add` migration moves top-level account-scoped keys into `accounts.default`. For channels that declare `namedAccountPromotionKeys`, this is limited to credentials only — policy keys stay at the channel top level as shared default... | open | null | false | 1 | [] | [] | 2026-04-07T08:11:29Z | 2026-05-02T07:02:10Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | alexey-pelykh | 345,102 | MDQ6VXNlcjM0NTEwMg== | User | false |
openclaw/openclaw | 4,217,962,045 | I_kwDOQb6kR877aP49 | 62,514 | https://github.com/openclaw/openclaw/issues/62514 | https://api.github.com/repos/openclaw/openclaw/issues/62514 | [Bug]: WEBUI / Edge cases —— Image input causes some multimodal models to misuse tool calling | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
In the Web UI, image input can cause some multimodal models with weak tool-calling capabilities to misuse tool calls, which may result in the image input being ignored or hallucinations in image understanding.
... | open | null | false | 2 | [
"bug"
] | [] | 2026-04-07T13:23:58Z | 2026-05-02T06:31:29Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | xiaoHEI-312 | 44,249,876 | MDQ6VXNlcjQ0MjQ5ODc2 | User | false |
openclaw/openclaw | 4,213,693,551 | I_kwDOQb6kR877J9xv | 62,095 | https://github.com/openclaw/openclaw/issues/62095 | https://api.github.com/repos/openclaw/openclaw/issues/62095 | Post-upgrade stability regressions — v2026.4.5 (3e72c03) | **Post-upgrade stability issues — v2026.4.5 (3e72c03)**
Upgraded to 2026.4.5 this morning. System was stable before the upgrade. Experienced 10 gateway restarts in ~8 hours due to several issues. Environment: Mac Studio M3 Ultra, BlueBubbles iMessage, local loopback gateway.
**1. `doctor --fix` doesn't fix its own wa... | open | null | false | 4 | [] | [] | 2026-04-06T19:36:54Z | 2026-05-02T06:09:43Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | jmnickels | 3,698,642 | MDQ6VXNlcjM2OTg2NDI= | User | false |
openclaw/openclaw | 4,211,622,077 | I_kwDOQb6kR877CEC9 | 61,864 | https://github.com/openclaw/openclaw/issues/61864 | https://api.github.com/repos/openclaw/openclaw/issues/61864 | compaction can immediately re-enter on stale totals when no trustworthy post-compaction snapshot is persisted | ## Summary
After auto memory-flush compaction, the next turn can immediately re-enter compaction if the session store does not persist a trustworthy post-compaction token snapshot or explicitly invalidate stale totals.
This is not just a `/status` display issue. It affects the actual compaction gate.
## User-visible... | open | null | false | 1 | [] | [] | 2026-04-06T12:28:44Z | 2026-05-02T07:02:07Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | dashhuang | 125,997,726 | U_kgDOB4KSng | User | false |
openclaw/openclaw | 4,216,839,022 | I_kwDOQb6kR877V9tu | 62,416 | https://github.com/openclaw/openclaw/issues/62416 | https://api.github.com/repos/openclaw/openclaw/issues/62416 | Accessibility Improvement: Control UI Settings interface is not compatible with screen readers | Hi OpenClaw Team,
I would like to report a significant accessibility issue regarding the Control UI. Currently, the settings interface is not fully compatible with screen readers used by visually impaired users.
The navigation and organization of the settings panel make it difficult for screen reader users to effici... | closed | not_planned | true | 1 | [] | [] | 2026-04-07T09:46:09Z | 2026-05-03T04:34:35Z | 2026-05-01T00:08:36Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | aim9sour | 214,139,338 | U_kgDODMOByg | User | false |
openclaw/openclaw | 4,213,098,355 | I_kwDOQb6kR877Hsdz | 62,042 | https://github.com/openclaw/openclaw/issues/62042 | https://api.github.com/repos/openclaw/openclaw/issues/62042 | Slack proactive DM delivery fails while direct chat.postMessage to user ID succeeds | ## Summary
Slack proactive DM delivery appears broken in OpenClaw's delivery path, even though the underlying Slack app can successfully send a DM directly via `chat.postMessage` to the same user.
## Expected
When a Slack bot can successfully send a DM using `chat.postMessage` with `channel=<USER_ID>`, OpenClaw should... | closed | completed | true | 2 | [] | [] | 2026-04-06T17:36:32Z | 2026-05-04T04:33:58Z | 2026-05-02T04:16:36Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | MarkMolina | 3,329,668 | MDQ6VXNlcjMzMjk2Njg= | User | false |
openclaw/openclaw | 4,209,368,874 | I_kwDOQb6kR8765d8q | 61,661 | https://github.com/openclaw/openclaw/issues/61661 | https://api.github.com/repos/openclaw/openclaw/issues/61661 | Control UI restores historical routed session instead of main session after reload | Hi, I'm seeing what looks like a Control UI session restore issue and wanted to ask whether this is the intended behavior.
## What I observe
My current inbound context is from `webchat` / `openclaw-control-ui`, and the actual heartbeat target is still the main session.
However, after reopening or reloading the Contr... | open | null | false | 1 | [] | [] | 2026-04-06T03:32:18Z | 2026-05-02T05:56:21Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | junoclaws | 266,291,263 | U_kgDOD99IPw | User | false |
openclaw/openclaw | 4,358,117,993 | I_kwDOQb6kR88AAAABA8OaaQ | 75,077 | https://github.com/openclaw/openclaw/issues/75077 | https://api.github.com/repos/openclaw/openclaw/issues/75077 | [Bug]: [2026.4.26] Tool call intermediate output bypasses SOUL.md output filter rules | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Tool call intermediate steps bypass SOUL.md output filter rules after upgrade to 2026.4.26
### Steps to reproduce
1. Configure SOUL.md with output filter rules (Section 4 and final section)
2. Send any message that trigger... | closed | completed | true | 3 | [
"bug"
] | [] | 2026-04-30T12:34:43Z | 2026-05-03T04:37:35Z | 2026-04-30T12:39:31Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | googlerest | 127,843,198 | U_kgDOB567fg | User | false |
openclaw/openclaw | 4,208,046,864 | I_kwDOQb6kR8760bMQ | 61,432 | https://github.com/openclaw/openclaw/issues/61432 | https://api.github.com/repos/openclaw/openclaw/issues/61432 | Bug: event.messages.push() on message:received has no effect | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
event.messages.push() executes without error on message:received events but has no observable effect in any channel (webchat or Telegram), despite the docs stating messages is available to push messages to the ... | closed | not_planned | true | 2 | [
"bug",
"bug:behavior"
] | [] | 2026-04-05T16:35:02Z | 2026-05-03T04:37:33Z | 2026-04-30T13:08:53Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | johnrobertcobbold | 307,565 | MDQ6VXNlcjMwNzU2NQ== | User | false |
openclaw/openclaw | 4,214,810,642 | I_kwDOQb6kR877OOgS | 62,212 | https://github.com/openclaw/openclaw/issues/62212 | https://api.github.com/repos/openclaw/openclaw/issues/62212 | Vertex AI ADC OAuth Authentication Fails with 401 Error | ## Problem Description
When using Vertex AI with Application Default Credentials (ADC), OpenClaw returns 401 status code (no body) error. Direct API calls with curl using the same ADC credentials work correctly.
## Environment
- OpenClaw version: 2026.4.2
- pi-ai version: 0.64.0
- @google/genai version: 1.47.0
- GCP... | closed | completed | true | 1 | [] | [] | 2026-04-07T00:52:39Z | 2026-05-03T04:37:33Z | 2026-04-30T13:09:01Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | nakedsleepwalker | 267,792,147 | U_kgDOD_YvEw | User | false |
openclaw/openclaw | 4,207,496,073 | I_kwDOQb6kR876yUuJ | 61,347 | https://github.com/openclaw/openclaw/issues/61347 | https://api.github.com/repos/openclaw/openclaw/issues/61347 | Feature: Selective/lazy injection for workspace context files | ## Problem
OpenClaw's `boot-md` hook injects all workspace context files (AGENTS.md, SOUL.md, IDENTITY.md, USER.md, TOOLS.md, HEARTBEAT.md, MEMORY.md) into every single turn — regardless of whether they're needed. This burns tokens on every message, and the cost compounds on expensive models like Opus.
After manually... | open | null | false | 1 | [] | [] | 2026-04-05T13:03:28Z | 2026-05-02T07:02:01Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | simonlpaige | 37,185,484 | MDQ6VXNlcjM3MTg1NDg0 | User | false |
openclaw/openclaw | 4,208,406,693 | I_kwDOQb6kR8761zCl | 61,496 | https://github.com/openclaw/openclaw/issues/61496 | https://api.github.com/repos/openclaw/openclaw/issues/61496 | Gateway fails to start when allowedSessionKeyPrefixes contains only sub-prefixes (e.g. hook:ao:) | ## Summary
When `hooks.allowedSessionKeyPrefixes` is set to a sub-prefix like `["hook:ao:"]` without a `defaultSessionKey`, the gateway refuses to start with:
```
Gateway failed to start: Error: hooks.allowedSessionKeyPrefixes must include 'hook:' when hooks.defaultSessionKey is unset
```
This is a usability issue b... | open | null | false | 1 | [] | [] | 2026-04-05T19:30:39Z | 2026-05-02T05:56:20Z | null | NONE | null | 20260502T060142Z | 2026-05-02T06:01:42Z | priyanshuharshbodhi1 | 132,090,159 | U_kgDOB9-JLw | User | false |
openclaw/openclaw | 4,209,120,591 | I_kwDOQb6kR8764hVP | 61,616 | https://github.com/openclaw/openclaw/issues/61616 | https://api.github.com/repos/openclaw/openclaw/issues/61616 | [Bug]: [WSL2] Global 30-min gateway stall (:29/:59) affects Telegram + Control UI | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
## Summary
Gateway stalls for ~2–3 minutes at exact half-hour boundaries (:29-:32 and :59-:02).
During stall, Telegram and Control UI both become unresponsive.
## Environment
- OpenClaw: 2026.4.2 (d74a122)
- OS: WSL2 Ubuntu... | open | null | false | 2 | [
"bug"
] | [] | 2026-04-06T02:14:36Z | 2026-05-02T07:02:03Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | alexandrey128-cell | 264,919,392 | U_kgDOD8pZYA | User | false |
openclaw/openclaw | 4,209,253,030 | I_kwDOQb6kR8765Bqm | 61,649 | https://github.com/openclaw/openclaw/issues/61649 | https://api.github.com/repos/openclaw/openclaw/issues/61649 | Google/Gmail operability recovery is too fragile to serve as a sustainable baseline | ## Summary
Google/Gmail/Drive operability on a live OpenClaw instance is too fragile to treat as a sustainable baseline. The capability exists and has worked, but recovery repeatedly depends on brittle interactive OAuth/keyring/callback flows that are not operationally closed for a production-like instance.
## Why thi... | open | null | false | 1 | [] | [] | 2026-04-06T02:59:30Z | 2026-05-02T07:02:06Z | null | NONE | null | 20260502T090030Z | 2026-05-02T09:00:30Z | tadeogutierrez | 16,327,260 | MDQ6VXNlcjE2MzI3MjYw | User | false |
openclaw/openclaw | 4,358,321,408 | I_kwDOQb6kR88AAAABA8a1AA | 75,086 | https://github.com/openclaw/openclaw/issues/75086 | https://api.github.com/repos/openclaw/openclaw/issues/75086 | [Audit] Theme Factory color accuracy: brand-inspired themes use wrong colors | # Theme Factory Color Audit Report
## Executive Summary
This issue documents a comprehensive color audit of the `theme-factory` skill's 10 preset themes, plus proposed additions. Multiple themes contain **factually incorrect color assignments** that misrepresent the brand identities they claim to emulate. We propose ... | closed | not_planned | true | 1 | [] | [] | 2026-04-30T13:06:14Z | 2026-05-03T04:37:32Z | 2026-04-30T13:17:59Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | 255doesnotexist | 19,223,209 | MDQ6VXNlcjE5MjIzMjA5 | User | false |
openclaw/openclaw | 4,358,451,615 | I_kwDOQb6kR88AAAABA8ixnw | 75,091 | https://github.com/openclaw/openclaw/issues/75091 | https://api.github.com/repos/openclaw/openclaw/issues/75091 | openclaw config set strips unrelated top-level fields (gateway block disappears, exit 78/CONFIG) | ## Bug: `openclaw config set <key>` strips unrelated top-level fields on every write
### Summary
Calling `openclaw config set <a.b.c> "<value>"` reads `~/.openclaw/openclaw.json`, applies the change, and writes it back through a schema-aware serializer that **silently drops top-level fields it doesn't recognize for t... | closed | not_planned | true | 1 | [] | [] | 2026-04-30T13:24:11Z | 2026-05-03T04:37:31Z | 2026-04-30T13:29:49Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | fahedyasin | 5,945,904 | MDQ6VXNlcjU5NDU5MDQ= | User | false |
openclaw/openclaw | 4,337,865,362 | I_kwDOQb6kR88AAAABAo6Skg | 72,968 | https://github.com/openclaw/openclaw/issues/72968 | https://api.github.com/repos/openclaw/openclaw/issues/72968 | [Bug]: imageModel config not intercepting incoming message attachments as documented | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Attachments are dropped with "model does not support images" warnings despite a valid imageModel (Gemini 2.0 Flash) being configured to handle vision processing when the primary model (Tencent Hy3) lacks image ... | closed | completed | true | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-04-27T18:03:02Z | 2026-05-03T04:37:27Z | 2026-04-30T13:50:06Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | vorajeeah | 144,963,184 | U_kgDOCKP2cA | User | false |
openclaw/openclaw | 4,358,601,818 | I_kwDOQb6kR88AAAABA8r8Wg | 75,098 | https://github.com/openclaw/openclaw/issues/75098 | https://api.github.com/repos/openclaw/openclaw/issues/75098 | v2026.4.27 group reply default change causes silent breakage for Slack/Telegram channel users | ## Summary
Upgrading from v2026.4.26 to v2026.4.27 silently stops the bot from replying to mentions in Slack channels (and likely Discord/Telegram/Matrix groups) **without any user-facing warning**. The reply still appears in the Control UI / webchat transcript, but is never delivered to the originating channel.
## R... | closed | not_planned | true | 1 | [] | [] | 2026-04-30T13:46:17Z | 2026-05-03T04:37:22Z | 2026-04-30T13:51:37Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | EthanChouTW | 11,582,478 | MDQ6VXNlcjExNTgyNDc4 | User | false |
openclaw/openclaw | 4,358,552,631 | I_kwDOQb6kR88AAAABA8o8Nw | 75,096 | https://github.com/openclaw/openclaw/issues/75096 | https://api.github.com/repos/openclaw/openclaw/issues/75096 | [Bug]: Yuanbao group chat: model output routed to wrong session (direct queue instead of group queue) — replies never sent to group | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Group chat @mention triggers model generation but the reply is queued in the direct session's outbound buffer (agent:main:yuanbao:direct:...) instead of the group session's buffer (agent:main:yuanbao:group:387668713), causin... | closed | completed | true | 1 | [
"bug",
"regression"
] | [] | 2026-04-30T13:39:05Z | 2026-05-03T04:37:19Z | 2026-04-30T13:58:56Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | kingsword520 | 63,271,778 | MDQ6VXNlcjYzMjcxNzc4 | User | false |
openclaw/openclaw | 4,208,144,610 | I_kwDOQb6kR8760zDi | 61,445 | https://github.com/openclaw/openclaw/issues/61445 | https://api.github.com/repos/openclaw/openclaw/issues/61445 | [Bug]: iMessage echo loop on gateway restart — echo cache not warmed, self-messages replay as inbound | ## Summary
After a gateway restart, the in-memory echo cache is empty. iMessage outbound replies written to the Messages DB with `is_from_me=True` and `sender=<bot_number>` are not matched by the echo filter, so they slip through as new inbound messages — triggering a reply loop.
## Steps to Reproduce
1. Have the iM... | closed | not_planned | true | 1 | [] | [] | 2026-04-05T17:20:22Z | 2026-05-03T04:37:17Z | 2026-04-30T13:59:49Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | loki-mamv | 258,358,405 | U_kgDOD2Y8hQ | User | false |
openclaw/openclaw | 4,208,467,496 | I_kwDOQb6kR8762B4o | 61,500 | https://github.com/openclaw/openclaw/issues/61500 | https://api.github.com/repos/openclaw/openclaw/issues/61500 | [Feature]: add system.read / system.write to node protocol for remote file I/O | ### Summary
Add system.read and system.write commands to the node protocol so that file I/O tools (read, write, edit) can route to the node host when tools.exec.host=node, enabling a coding agent to read and write files on the same machine where it executes code.
### Problem to solve
Problem
When using a two-machine... | closed | completed | true | 1 | [
"enhancement"
] | [] | 2026-04-05T20:03:17Z | 2026-05-03T04:37:16Z | 2026-04-30T13:59:55Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | AndrewJWicks | 177,733,750 | U_kgDOCpgAdg | User | false |
openclaw/openclaw | 4,211,622,164 | I_kwDOQb6kR877CEEU | 61,865 | https://github.com/openclaw/openclaw/issues/61865 | https://api.github.com/repos/openclaw/openclaw/issues/61865 | `/status` can report stale pre-compaction totals for already-compacted sessions | ## Summary
For already-compacted sessions, `/status` can still report stale pre-compaction context totals (for example `151k/128k`) even after the live prompt has already been reduced.
This appears to happen when status rendering prefers an older store snapshot or a whole-transcript estimate that still reflects the p... | closed | not_planned | true | 1 | [] | [] | 2026-04-06T12:28:45Z | 2026-05-03T04:37:15Z | 2026-04-30T14:00:01Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | dashhuang | 125,997,726 | U_kgDOB4KSng | User | false |
openclaw/openclaw | 4,337,145,100 | I_kwDOQb6kR88AAAABAoOVDA | 72,903 | https://github.com/openclaw/openclaw/issues/72903 | https://api.github.com/repos/openclaw/openclaw/issues/72903 | [Bug]: New inbound message during deferred gateway restart leaves main session stuck in processing (restart-deferred deadlock) | ## Bug type
Behavior bug (incorrect output/state without crash) — runtime deadlock
## Beta release blocker
No
## Summary
When `config.apply` triggers a gateway restart that gets deferred until active task runs drain, any new inbound message that lands in that deferral window opens a brand-new task run. Since `rest... | closed | completed | true | 1 | [] | [] | 2026-04-27T16:08:10Z | 2026-05-03T04:35:10Z | 2026-04-30T19:53:54Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | chenyanchen | 23,415,847 | MDQ6VXNlcjIzNDE1ODQ3 | User | false |
openclaw/openclaw | 4,358,732,832 | I_kwDOQb6kR88AAAABA8z8IA | 75,105 | https://github.com/openclaw/openclaw/issues/75105 | https://api.github.com/repos/openclaw/openclaw/issues/75105 | [Feature]: Allow per-model setting for Deepseek `reasoning_content` fix | ### Summary
Proposal to allow per-model configuration for `reasoning_content` fix for Deepseek v4 or any other (future) models that might require it, which is also required right now for any API providers that offer DeepSeek Model access.
### Problem to solve
Recent fixes (https://github.com/openclaw/openclaw/issue... | open | null | false | 3 | [
"enhancement"
] | [] | 2026-04-30T14:05:30Z | 2026-05-03T10:41:49Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | flofrie | 47,421,818 | MDQ6VXNlcjQ3NDIxODE4 | User | false |
openclaw/openclaw | 4,358,875,726 | I_kwDOQb6kR88AAAABA88qTg | 75,112 | https://github.com/openclaw/openclaw/issues/75112 | https://api.github.com/repos/openclaw/openclaw/issues/75112 | Feature Request: 飞书频道流式卡片展示工具调用过程 | null | closed | not_planned | true | 1 | [] | [] | 2026-04-30T14:26:37Z | 2026-05-03T04:37:00Z | 2026-04-30T14:32:32Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | redhill150429-ui | 270,263,671 | U_kgDOEBvldw | User | false |
openclaw/openclaw | 4,337,235,944 | I_kwDOQb6kR88AAAABAoT36A | 72,908 | https://github.com/openclaw/openclaw/issues/72908 | https://api.github.com/repos/openclaw/openclaw/issues/72908 | [Bug]: `openclaw update --tag 2026.4.25` can leave the old Gateway running against replaced package files, causing mixed-version runtime/import failures |
## Summary
Updating from OpenClaw `2026.4.24` to `2026.4.25` with the documented CLI update path completed at the package level, but the updater exited with code `1` and no stdout/stderr. Afterward, `openclaw --version` reported `2026.4.25`, while the already-running Gateway process still reported runtime `2026.4.24... | closed | completed | true | 1 | [] | [] | 2026-04-27T16:22:00Z | 2026-05-03T04:36:57Z | 2026-04-30T14:32:50Z | CONTRIBUTOR | null | 20260503T060030Z | 2026-05-03T06:00:30Z | vishutdhar | 68,405,187 | MDQ6VXNlcjY4NDA1MTg3 | User | false |
openclaw/openclaw | 4,358,881,290 | I_kwDOQb6kR88AAAABA89ACg | 75,113 | https://github.com/openclaw/openclaw/issues/75113 | https://api.github.com/repos/openclaw/openclaw/issues/75113 | Feature Request: 飞书频道流式卡片展示工具调用过程 | ## Feature Request: 飞书频道流式卡片展示工具调用过程
### 背景
当前飞书频道的流式卡片只能展示两类内容：
- 思考过程（reasoning text）
- 最终回复（response text）
在 reasoning: medium 模式下，用户看不到任何工作过程——模型在后台跑了几十秒，用户盯着空白等待。
### 期望行为
类似 Hermes 的效果：流式卡片中实时展示当前正在执行的工具调用，例如：
⏳ Thinking...
🔧 正在读取文件: /path/to/config.json
✅ 读取完成 (120ms)
🔧 正在执行命令: grep "streaming" config.js... | closed | completed | true | 1 | [] | [] | 2026-04-30T14:27:26Z | 2026-05-03T04:36:57Z | 2026-04-30T14:33:44Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | redhill150429-ui | 270,263,671 | U_kgDOEBvldw | User | false |
openclaw/openclaw | 4,358,368,287 | I_kwDOQb6kR88AAAABA8dsHw | 75,087 | https://github.com/openclaw/openclaw/issues/75087 | https://api.github.com/repos/openclaw/openclaw/issues/75087 | [Bug]: Browser runtime ignores configured executablePath (regression, original #73617 bot-closed prematurely) | **This is a follow-up to #73617, which was closed prematurely by a bot (clawsweeper[bot]) claiming the fix was on `main`, despite the issue persisting in the latest release.**
## Summary
The OpenClaw browser runtime ignores the configured `browser.executablePath` (and `browser.profiles.openclaw.executablePath`) in th... | closed | completed | true | 2 | [] | [] | 2026-04-30T13:12:47Z | 2026-05-03T04:36:55Z | 2026-04-30T14:36:29Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | civiltox | 97,610,456 | U_kgDOBdFq2A | User | false |
openclaw/openclaw | 4,358,017,882 | I_kwDOQb6kR88AAAABA8ITWg | 75,074 | https://github.com/openclaw/openclaw/issues/75074 | https://api.github.com/repos/openclaw/openclaw/issues/75074 | [Feature]: /v1/responses drops every built-in tool call from `output`; add opt-in flag to surface them | ## Three concrete things you cannot do today
1. **Run an offline eval against `/v1/responses` that scores tool use.**
Sample prompt: "find every TODO in the repo and write a summary." We want to grade the agent on whether it called `grep`, `find`, and `read` in a sensible order — without scraping logs out-of-band. ... | open | null | false | 4 | [] | [] | 2026-04-30T12:20:28Z | 2026-05-03T08:40:20Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | glow1128 | 5,348,506 | MDQ6VXNlcjUzNDg1MDY= | User | false |
openclaw/openclaw | 4,358,306,338 | I_kwDOQb6kR88AAAABA8Z6Ig | 75,084 | https://github.com/openclaw/openclaw/issues/75084 | https://api.github.com/repos/openclaw/openclaw/issues/75084 | [Bug]: echoTranscript works in Telegram forum group but not in direct chat (DM) | ## Summary
The `echoTranscript` feature delivers the transcript echo correctly in Telegram forum supergroups, but fails to deliver it in Telegram direct chats (DMs). Transcription itself works in both contexts — only the echo message delivery is affected.
## Environment
- OpenClaw version: v2026.4.27 (stable)
- Chan... | closed | completed | true | 3 | [
"dedupe:parent"
] | [] | 2026-04-30T13:03:57Z | 2026-05-03T04:33:44Z | 2026-04-30T14:45:49Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | M-Lietz | 222,629,290 | U_kgDODUUNqg | User | false |
openclaw/openclaw | 4,358,379,474 | I_kwDOQb6kR88AAAABA8eX0g | 75,088 | https://github.com/openclaw/openclaw/issues/75088 | https://api.github.com/repos/openclaw/openclaw/issues/75088 | web_search tool description says Brave API even when DuckDuckGo is active | ### Bug type
Tool metadata / provider description mismatch
### Summary
In a session where Brave Search is not configured, `web_search` correctly uses DuckDuckGo, but the assistant-facing tooling description says:
> web_search: Search the web (Brave API)
This is misleading. It caused the assistant to incorrectly an... | closed | completed | true | 3 | [] | [] | 2026-04-30T13:14:21Z | 2026-05-03T04:36:55Z | 2026-04-30T14:46:06Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | sun-rongyang | 20,738,314 | MDQ6VXNlcjIwNzM4MzE0 | User | false |
openclaw/openclaw | 4,358,976,301 | I_kwDOQb6kR88AAAABA9CzLQ | 75,115 | https://github.com/openclaw/openclaw/issues/75115 | https://api.github.com/repos/openclaw/openclaw/issues/75115 | openclaw gateway restart + systemd Restart=always = EADDRINUSE spawn loop | ## Bug
When `openclaw gateway restart` is used with a systemd service configured with `Restart=always`, it causes a spawn loop of zombie gateway processes that burn CPU.
## Setup
- OpenClaw v2026.4.26
- systemd user service with `Restart=always`, `RestartSec=5`, `TimeoutStopSec=30`
## What happens
1. `openclaw gate... | closed | completed | true | 2 | [] | [] | 2026-04-30T14:40:44Z | 2026-05-03T04:36:17Z | 2026-04-30T15:31:23Z | CONTRIBUTOR | null | 20260503T060030Z | 2026-05-03T06:00:30Z | yhyatt | 10,474,956 | MDQ6VXNlcjEwNDc0OTU2 | User | false |
openclaw/openclaw | 4,358,955,789 | I_kwDOQb6kR88AAAABA9BjDQ | 75,114 | https://github.com/openclaw/openclaw/issues/75114 | https://api.github.com/repos/openclaw/openclaw/issues/75114 | [Bug]: Telegram polling creates new HTTPS connection every ~10 seconds instead of reusing keepalive connections | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
## Environment
- **OpenClaw version:** 2026.4.27
- **macOS:** 26.3.1 (Darwin 25.3.0 arm64)
- **Gateway PID:** 21935
- **Node:** v25.8.1
- **Date:** 2026-04-30
## Steps to Reproduce
1. Run `openclaw gateway start`
2. Monito... | closed | completed | true | 2 | [
"bug"
] | [] | 2026-04-30T14:37:45Z | 2026-05-03T04:36:22Z | 2026-04-30T15:14:28Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | hpinho77 | 3,128,540 | MDQ6VXNlcjMxMjg1NDA= | User | false |
openclaw/openclaw | 4,334,358,665 | I_kwDOQb6kR88AAAABAlkQiQ | 72,717 | https://github.com/openclaw/openclaw/issues/72717 | https://api.github.com/repos/openclaw/openclaw/issues/72717 | Feature Request: SQLite FTS index for wiki_search to improve synthesis query performance | ## Is your feature request related to a problem?
When the wiki vault grows (80+ synthesis pages), `wiki_search` scans Markdown files for matches. This works but becomes increasingly inefficient as an agent's knowledge base scales. Each search requires reading and parsing multiple `.md` files, consuming tool-call token... | open | null | false | 2 | [] | [] | 2026-04-27T08:45:29Z | 2026-05-03T10:41:34Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | MagnaAI | 265,843,828 | U_kgDOD9h0dA | User | false |
openclaw/openclaw | 4,336,856,692 | I_kwDOQb6kR88AAAABAn8udA | 72,875 | https://github.com/openclaw/openclaw/issues/72875 | https://api.github.com/repos/openclaw/openclaw/issues/72875 | `memorySearch provider: "local"` fails with "Unknown memory embedding provider: local" but capability embedding path works | ## Summary
`memorySearch.provider = "local"` appears to be a supported memory embedding configuration, but the memory CLI/runtime path can fail with:
```text
Unknown memory embedding provider: local
```
The same local provider works through the embedding capability CLI, so this may be a provider-registration differe... | open | null | false | 1 | [] | [] | 2026-04-27T15:21:19Z | 2026-05-03T07:05:12Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | Amirofcodes | 138,374,972 | U_kgDOCD9vPA | User | false |
openclaw/openclaw | 4,336,955,296 | I_kwDOQb6kR88AAAABAoCvoA | 72,883 | https://github.com/openclaw/openclaw/issues/72883 | https://api.github.com/repos/openclaw/openclaw/issues/72883 | gateway config.patch blocks `diagnostics.cacheTrace.*` even with content capture disabled | ## Summary
`gateway.config.patch` (and the underlying protected-paths list) refuses to set `diagnostics.cacheTrace.enabled` even when the patch also explicitly disables all sensitive content capture (`includeMessages: false`, `includePrompt: false`, `includeSystem: false`).
The schema help text for `diagnostics.cacheT... | open | null | false | 1 | [] | [] | 2026-04-27T15:37:35Z | 2026-05-03T09:13:12Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | jessewunderlich | 83,950,240 | MDQ6VXNlcjgzOTUwMjQw | User | false |
openclaw/openclaw | 4,336,955,298 | I_kwDOQb6kR88AAAABAoCvog | 72,884 | https://github.com/openclaw/openclaw/issues/72884 | https://api.github.com/repos/openclaw/openclaw/issues/72884 | `tasks maintenance --apply` is a no-op for `lost` (backing-session-missing) and `stale_running` task records | ## Summary
`openclaw tasks maintenance --apply` reports `0 reconcile · 0 recovered · 0 cleanup stamp · 0 prune` even when `openclaw tasks audit` reports `lost` and `stale_running` errors. These records are exactly what an operator would expect maintenance to clean.
## Reproduction
On a long-lived install with the defa... | open | null | false | 2 | [] | [] | 2026-04-27T15:37:35Z | 2026-05-03T07:05:14Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | jessewunderlich | 83,950,240 | MDQ6VXNlcjgzOTUwMjQw | User | false |
openclaw/openclaw | 4,337,003,828 | I_kwDOQb6kR88AAAABAoFtNA | 72,887 | https://github.com/openclaw/openclaw/issues/72887 | https://api.github.com/repos/openclaw/openclaw/issues/72887 | Dreaming deep promotion should support a review queue before writing to MEMORY.md | ## Summary
`memory-core` Dreaming currently writes deep-phase promoted candidates directly into `MEMORY.md` as `## Promoted From Short-Term Memory (...)` blocks. I would like maintainers' guidance on whether this final write path should be configurable so promoted output can land in a separate review queue first, inst... | open | null | false | 1 | [] | [] | 2026-04-27T15:44:57Z | 2026-05-03T08:39:42Z | null | CONTRIBUTOR | null | 20260503T090142Z | 2026-05-03T09:01:42Z | solavrc | 145,330,217 | U_kgDOCKmQKQ | User | false |
openclaw/openclaw | 4,337,011,122 | I_kwDOQb6kR88AAAABAoGJsg | 72,889 | https://github.com/openclaw/openclaw/issues/72889 | https://api.github.com/repos/openclaw/openclaw/issues/72889 | [Bug]: pi-ai anthropic adapter does not merge consecutive same-role messages — causes 400 errors with custom proxies | ## Bug type
Missing message normalization in `@mariozechner/pi-ai` anthropic provider adapter.
## Summary
The `convertMessages()` function in `@mariozechner/pi-ai/dist/providers/anthropic.js` does not merge consecutive messages with the same `role` before sending to the Anthropic API. The Anthropic Messages API requi... | open | null | false | 2 | [] | [] | 2026-04-27T15:46:08Z | 2026-05-03T10:51:15Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | fuyizheng3120 | 268,386,865 | U_kgDOD_9CMQ | User | false |
openclaw/openclaw | 4,337,013,103 | I_kwDOQb6kR88AAAABAoGRbw | 72,890 | https://github.com/openclaw/openclaw/issues/72890 | https://api.github.com/repos/openclaw/openclaw/issues/72890 | [Feature]: supportsAdaptiveThinking should respect custom provider model capabilities | ## Feature type
Enhancement to adaptive thinking model detection.
## Summary
`supportsAdaptiveThinking()` in `@mariozechner/pi-ai/dist/providers/anthropic.js` uses a hardcoded model ID allowlist (`opus-4-6`, `opus-4.6`, `opus-4-7`, `sonnet-4-6`, etc.). Custom providers that proxy Anthropic models under different mode... | open | null | false | 2 | [] | [] | 2026-04-27T15:46:26Z | 2026-05-03T10:51:15Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | fuyizheng3120 | 268,386,865 | U_kgDOD_9CMQ | User | false |
openclaw/openclaw | 4,337,048,578 | I_kwDOQb6kR88AAAABAoIcAg | 72,893 | https://github.com/openclaw/openclaw/issues/72893 | https://api.github.com/repos/openclaw/openclaw/issues/72893 | DeepSeek provider plugin missing resolveThinkingProfile — /think max/xhigh rejected on deepseek-v4-pro despite full backend support | ## Summary
The bundled `deepseek` provider plugin in OpenClaw v2026.4.25 stable does **not** register a `resolveThinkingProfile()` hook, which causes `/think max` and `/think xhigh` to be rejected for `deepseek/deepseek-v4-pro` and `deepseek/deepseek-v4-flash`, even though:
1. **DeepSeek's official API supports both*... | closed | completed | true | 2 | [] | [] | 2026-04-27T15:52:37Z | 2026-05-03T04:36:14Z | 2026-04-30T15:34:07Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | superandylin | 56,207,675 | MDQ6VXNlcjU2MjA3Njc1 | User | false |
openclaw/openclaw | 4,337,055,810 | I_kwDOQb6kR88AAAABAoI4Qg | 72,895 | https://github.com/openclaw/openclaw/issues/72895 | https://api.github.com/repos/openclaw/openclaw/issues/72895 | WhatsApp outbound media upload can crash gateway with ENOENT on Baileys *-enc temp files | ### Summary
WhatsApp outbound media sends can fail with an uncaught `ENOENT` while Baileys is opening its temporary encrypted media file (`image...-enc` / `document...-enc`). In my case this repeatedly crashed/restarted the OpenClaw Gateway.
Text send and inbound WhatsApp media receive were working. The failure was s... | closed | completed | true | 2 | [] | [] | 2026-04-27T15:53:52Z | 2026-05-04T04:35:20Z | 2026-05-01T21:40:58Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | AnakinDang | 99,165,350 | U_kgDOBekkpg | User | false |
openclaw/openclaw | 4,337,396,174 | I_kwDOQb6kR88AAAABAodpzg | 72,927 | https://github.com/openclaw/openclaw/issues/72927 | https://api.github.com/repos/openclaw/openclaw/issues/72927 | feat(tts): support MiMo v2.5 TTS VoiceDesign | ## Feature request
Please add support for Xiaomi MiMo v2.5 TTS `VoiceDesign` in OpenClaw's TTS provider configuration.
## Background
OpenClaw can already use the Xiaomi TTS provider with `mimo-v2.5-tts` and a fixed `voice` value, for example:
```json
{
"messages": {
"tts": {
"auto": "tagged",
"pro... | open | null | false | 1 | [] | [] | 2026-04-27T16:47:26Z | 2026-05-03T03:37:36Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | Lunareith | 262,595,192 | U_kgDOD6bieA | User | false |
openclaw/openclaw | 4,337,507,133 | I_kwDOQb6kR88AAAABAokbPQ | 72,934 | https://github.com/openclaw/openclaw/issues/72934 | https://api.github.com/repos/openclaw/openclaw/issues/72934 | [Bug]: Browser snapshot action consistently times out despite successful open/navigate | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
OpenClaw Browser Automation Bug Report
Issue Summary: Browser automation consistently times out on snapshot action after successful open/navigate operations.
Environment:
• OpenClaw Version: 2026.4.25 (aa36ee6)
• OS: macO... | open | null | false | 1 | [
"bug",
"regression"
] | [] | 2026-04-27T17:05:11Z | 2026-05-03T11:49:55Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | wkomers-hash | 275,862,029 | U_kgDOEHFSDQ | User | false |
openclaw/openclaw | 4,337,414,656 | I_kwDOQb6kR88AAAABAoeyAA | 72,931 | https://github.com/openclaw/openclaw/issues/72931 | https://api.github.com/repos/openclaw/openclaw/issues/72931 | [Bug]: OpenClaw Control UI bootstrap blocked by CSP + Zone.js; gateway start/stop loops cause terminal flashing; no durable fix via CLI install options | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
Yes
### Summary
Bundled OpenClaw Control UI fails to bootstrap. The UI renders a blank/blue screen or restarts when the gateway is restarted via CLI stop/uninstall/install/start. Root cause appears to be a CSP in the gateway ... | closed | completed | true | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-04-27T16:50:10Z | 2026-05-03T04:36:43Z | 2026-04-30T14:54:15Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | jwilford99 | 102,714,853 | U_kgDOBh9N5Q | User | false |
openclaw/openclaw | 4,337,611,150 | I_kwDOQb6kR88AAAABAoqxjg | 72,952 | https://github.com/openclaw/openclaw/issues/72952 | https://api.github.com/repos/openclaw/openclaw/issues/72952 | [Bug]: dist/index.js subcommands in official Docker image take 24s-49s to start | ### Summary
`node dist/index.js` subcommands are extremely slow in the official Docker image even when run directly inside the running `openclaw-gateway` container.
This does **not** look like an Ansible issue. The timings below were measured directly on the machine with `docker compose exec`.
The same container exe... | open | null | false | 1 | [] | [] | 2026-04-27T17:21:54Z | 2026-05-03T01:38:11Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | shadow-identity | 1,223,112 | MDQ6VXNlcjEyMjMxMTI= | User | false |
openclaw/openclaw | 4,337,637,439 | I_kwDOQb6kR88AAAABAosYPw | 72,954 | https://github.com/openclaw/openclaw/issues/72954 | https://api.github.com/repos/openclaw/openclaw/issues/72954 | ForceSenderIsOwnerFalse=true for cron-fired announce delivery strips owner-only tools (cron, gateway) on next owner DM | ## Bug: `ForceSenderIsOwnerFalse=true` for cron-fired `delivery.mode: "announce"` strips owner-only tools (`cron`, `gateway`) for the next owner DM
### Summary
When an isolated cron job with `delivery.mode: "announce"` fires and delivers its output back into the originating Discord/Telegram DM session, the queued sys... | open | null | false | 1 | [] | [] | 2026-04-27T17:26:23Z | 2026-05-03T11:49:56Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | fychmen | 15,855,997 | MDQ6VXNlcjE1ODU1OTk3 | User | false |
openclaw/openclaw | 4,337,804,372 | I_kwDOQb6kR88AAAABAo2kVA | 72,964 | https://github.com/openclaw/openclaw/issues/72964 | https://api.github.com/repos/openclaw/openclaw/issues/72964 | Compaction emits empty fallback summary; tokensBefore counts cacheRead, triggering premature compactions on Opus 1M | # Compaction generates empty fallback summary; tokensBefore counter sums cache_read tokens, triggering compaction prematurely on Opus 1M
**Version**: 2026.4.24 (cbcfdf6)
**Severity**: High — silently destroys conversation context, breaking the
"compaction memory first" guarantee.
## Summary
Two related bugs observed... | open | null | false | 1 | [] | [] | 2026-04-27T17:52:33Z | 2026-05-03T07:50:09Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | mrossit | 207,357,822 | U_kgDODFwHfg | User | false |
openclaw/openclaw | 4,337,851,160 | I_kwDOQb6kR88AAAABAo5bGA | 72,966 | https://github.com/openclaw/openclaw/issues/72966 | https://api.github.com/repos/openclaw/openclaw/issues/72966 | [Bug]: openai-codex/gpt-5.5 hangs silently on 2026.4.25 — no tokens streamed, no error logged, legacy OAuth path | ### Bug type
Regression (worked before, now fails)
### Summary
On OpenClaw 2026.4.25, agent runs targeting `openai-codex/gpt-5.5` hang indefinitely on first turn. The model never streams a token, no error is logged, and the only way to recover is an external abort (stop button / external timeout). The trajectory art... | closed | completed | true | 1 | [] | [] | 2026-04-27T18:00:46Z | 2026-05-03T04:35:30Z | 2026-04-30T18:45:49Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | ped-ro | 12,043,963 | MDQ6VXNlcjEyMDQzOTYz | User | false |
openclaw/openclaw | 4,338,602,948 | I_kwDOQb6kR88AAAABApnTxA | 73,016 | https://github.com/openclaw/openclaw/issues/73016 | https://api.github.com/repos/openclaw/openclaw/issues/73016 | [Feature]: Native WAHA channel adapter (alongside Baileys) for stabler WhatsApp | ## Summary
Request native WAHA (https://github.com/devlikeapro/waha) as a first-class channel in OpenClaw, alongside the existing Baileys-backed `whatsapp` channel. WAHA is a self-hostable, Dockerized HTTP+webhook gateway around the WhatsApp Web protocol that abstracts the Baileys (`@whiskeysockets/baileys`) library b... | closed | not_planned | true | 1 | [] | [] | 2026-04-27T20:12:33Z | 2026-05-03T04:36:42Z | 2026-04-30T14:54:25Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | jared-rebel | 147,876,109 | U_kgDOCNBpDQ | User | false |
openclaw/openclaw | 4,338,393,624 | I_kwDOQb6kR88AAAABApaiGA | 73,003 | https://github.com/openclaw/openclaw/issues/73003 | https://api.github.com/repos/openclaw/openclaw/issues/73003 | [Bug]: Safeguard compaction permanently blocked by Pi SDK auto-compaction (consecutive failures) | # Bug Report: OpenClaw Safeguard Compaction Permanently Blocked by Pi SDK Auto-Compaction
## Summary
When `compaction.mode = "safeguard"` is configured, OpenClaw's preflight compaction is permanently blocked by the Pi SDK's internal auto-compaction. The two systems operate on different thresholds and do not coordinat... | open | null | false | 1 | [] | [] | 2026-04-27T19:36:01Z | 2026-05-03T07:05:21Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | bradhallett | 53,977,268 | MDQ6VXNlcjUzOTc3MjY4 | User | false |
openclaw/openclaw | 4,338,655,999 | I_kwDOQb6kR88AAAABApqi_w | 73,019 | https://github.com/openclaw/openclaw/issues/73019 | https://api.github.com/repos/openclaw/openclaw/issues/73019 | Proposal: Add xAI Realtime Voice Agent support to /voiceclaw/realtime via shared OpenAI-Realtime-protocol adapter | ## Summary
Add support for the **xAI Realtime Voice Agent** (`grok-voice-think-fast-1.0`) to OpenClaw's `/voiceclaw/realtime` gateway surface, implemented as a shared `OpenAIRealtimeProtocolAdapter` that can also serve OpenAI Realtime (issue #71195) when added.
xAI's Voice Agent API is OpenAI-Realtime-protocol-compat... | open | null | false | 1 | [] | [] | 2026-04-27T20:22:30Z | 2026-05-03T13:21:55Z | null | NONE | null | 20260503T150031Z | 2026-05-03T15:00:31Z | matthewtran172 | 257,667,908 | U_kgDOD1uzRA | User | false |
openclaw/openclaw | 4,338,795,047 | I_kwDOQb6kR88AAAABApzCJw | 73,035 | https://github.com/openclaw/openclaw/issues/73035 | https://api.github.com/repos/openclaw/openclaw/issues/73035 | [Bug] openclaw.runtime-context custom_message echoed verbatim in assistant responses (v2026.4.25) | ## Summary
After updating to OpenClaw 2026.4.25 (`aa36ee6`), the new `openclaw.runtime-context` custom message type is being echoed verbatim by agents as the first line of their responses. The runtime context preamble — intended as internal metadata — appears in user-visible output.
## Evidence
### Session transcrip... | closed | completed | true | 1 | [] | [] | 2026-04-27T20:49:24Z | 2026-05-03T04:36:41Z | 2026-04-30T14:54:32Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | SweetSophia | 44,297,511 | MDQ6VXNlcjQ0Mjk3NTEx | User | false |
openclaw/openclaw | 4,359,040,590 | I_kwDOQb6kR88AAAABA9GuTg | 75,116 | https://github.com/openclaw/openclaw/issues/75116 | https://api.github.com/repos/openclaw/openclaw/issues/75116 | [Feature] opt-in injectCallerContext for bundle MCP remote servers | ### Summary
Add an opt-in, per-server injectCallerContext flag on mcp.servers.<name> so OpenClaw can forward its caller-identity HTTP headers (x-openclaw-agent-id, x-openclaw-account-id, x-openclaw-message-channel, x-session-key) onto a remote MCP server when bundle MCP is enabled. The flag is off by default, only tak... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-30T14:49:57Z | 2026-05-03T14:00:30Z | null | NONE | null | 20260503T150031Z | 2026-05-03T15:00:31Z | david-garcia-garcia | 9,599,352 | MDQ6VXNlcjk1OTkzNTI= | User | false |
openclaw/openclaw | 4,358,776,355 | I_kwDOQb6kR88AAAABA82mIw | 75,108 | https://github.com/openclaw/openclaw/issues/75108 | https://api.github.com/repos/openclaw/openclaw/issues/75108 | Cross-version image upgrade: bundled-runtime-mirror EACCES on stale symlink in plugin-runtime-deps/ | ## Summary
When an OpenClaw container is upgraded to a new image version, gateway crash-loops with `EACCES: permission denied, copyfile … .openclaw-mirror-…tmp` because the bundled-runtime-mirror copies through stale symlinks pointing at the read-only image layer.
## Repro
Multi-tenant container deployment (badclaw ... | closed | completed | true | 3 | [] | [] | 2026-04-30T14:12:00Z | 2026-05-03T04:36:26Z | 2026-04-30T15:00:09Z | CONTRIBUTOR | null | 20260503T060030Z | 2026-05-03T06:00:30Z | coletebou | 12,384,893 | MDQ6VXNlcjEyMzg0ODkz | User | false |
openclaw/openclaw | 4,358,815,052 | I_kwDOQb6kR88AAAABA849TA | 75,109 | https://github.com/openclaw/openclaw/issues/75109 | https://api.github.com/repos/openclaw/openclaw/issues/75109 | Bug: collectAlreadyStagedBundledRuntimeDepSpecs omits auto-loaded plugins, causing their deps to be pruned | ## Summary
`collectAlreadyStagedBundledRuntimeDepSpecs` calls `collectBundledPluginRuntimeDeps({ extensionsDir, config })` with the user config. This causes auto-loaded bundled plugins that are **not** explicitly present in `config.plugins.entries` (e.g. `memory-core`) to be excluded from the dep scan. As a result the... | closed | completed | true | 3 | [] | [] | 2026-04-30T14:17:33Z | 2026-05-03T04:36:25Z | 2026-04-30T15:00:12Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | jared-rebel | 147,876,109 | U_kgDOCNBpDQ | User | false |
openclaw/openclaw | 4,357,974,990 | I_kwDOQb6kR88AAAABA8Frzg | 75,069 | https://github.com/openclaw/openclaw/issues/75069 | https://api.github.com/repos/openclaw/openclaw/issues/75069 | [Bug]: Bundled plugin runtime mirror runs synchronously on every pi-agent invocation, blocking the gateway main thread for tens of seconds (regression in 2026.4.22+) | ### Bug type
Regression
### Summary
Each pi-agent invocation triggers a synchronous "mirror" walk over every bundled plugin, blocking the gateway main thread. On the first agent run after gateway start, four full sweeps stack up and produce ~80–90 seconds of contiguous main-thread blocking (cold fs page cache). On s... | closed | completed | true | 5 | [] | [] | 2026-04-30T12:13:39Z | 2026-05-04T04:37:06Z | 2026-05-01T08:59:09Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | xiaohuaxi | 45,277,353 | MDQ6VXNlcjQ1Mjc3MzUz | User | false |
openclaw/openclaw | 4,359,188,165 | I_kwDOQb6kR88AAAABA9PuxQ | 75,123 | https://github.com/openclaw/openclaw/issues/75123 | https://api.github.com/repos/openclaw/openclaw/issues/75123 | Docs request: recommend apple-notes-native as a low-friction Apple Notes option on macOS | I’m opening this in the main OpenClaw repo rather than ClawHub because the request is now about product/docs guidance, not skill publishing.
Context:
- the Apple Notes skill has already been published to ClawHub: https://clawhub.ai/lishix520/apple-notes-native
- the earlier ClawHub discussion was closed as a third-par... | closed | not_planned | true | 1 | [] | [] | 2026-04-30T15:11:06Z | 2026-05-03T04:36:21Z | 2026-04-30T15:20:34Z | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | lishix520 | 36,358,469 | MDQ6VXNlcjM2MzU4NDY5 | User | false |
openclaw/openclaw | 4,058,294,933 | I_kwDOQb6kR87x5KqV | 43,220 | https://github.com/openclaw/openclaw/issues/43220 | https://api.github.com/repos/openclaw/openclaw/issues/43220 | [Feature]: Safer and More Reproducible Way to Share or Migrate OpenClaw Instances | ### Summary
<html><head></head><body><h2>Proposal / Discussion: Safer and More Reproducible Way to Share or Migrate OpenClaw Instances</h2><p>Hi OpenClaw team,</p><p>First of all, thanks for the amazing work on OpenClaw. I’ve been using it to build personal agents and workflows, and recently ran into some design quest... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-03-11T13:37:47Z | 2026-05-04T12:04:18Z | null | NONE | null | 20260504T180039Z | 2026-05-04T18:00:39Z | dreamcoin1998 | 47,708,051 | MDQ6VXNlcjQ3NzA4MDUx | User | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.