What makes OpenClaw Gateway Resilience Guard notable?

1.2K downloads. Trust evidence available

How should OpenClaw Gateway Resilience Guard be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

OpenClaw Gateway Resilience Guard

OpenClaw Gateway Resilience Guard keeps Gateway, channels, logs, and optional model-provider probes observable after Wi-Fi changes, WSL sleep/resume, session... Skill: OpenClaw Gateway Resilience Guard Owner: zc-kama Summary: OpenClaw Gateway Resilience Guard keeps Gateway, channels, logs, and optional model-provider probes observable after Wi-Fi changes, WSL sleep/resume, session... Tags: bash:1.2.0, channels:1.2.0, dashboard:1.4.4, gateway:1.4.4, guardian:1.2.0, latest:1.4.4, macos:1.4.4, openclaw:1.4.4, resilience:1.4.4, systemd:1.2.0, watchdog:1.4.4, wechat:1.4.4, weixin

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

Updated

Oct 11, 2026

Version

1.4.4

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.2K downloadsadoption · observed Oct 11, 2026
Latest release: 1.4.4release · observed May 21, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s1708vs1r95rbeb3za8tjcexfn871nt4:gateway-resilience-guard

Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-zc-kama-gateway-resilience-guard/snapshot"

Documentation

CLAWHUB

147,188 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: gateway-resilience-guard
description: OpenClaw Gateway Resilience Guard keeps Gateway, channels, logs, and optional model-provider probes observable after Wi-Fi changes, WSL sleep/resume, session expiry, provider timeouts, or partial outages. It adds a localhost dashboard with charts, event trends, language/theme switching, strategy presets, guarded action unlock, native OpenClaw health probes, restart safeguards, systemd, LaunchAgent, Task Scheduler, and an optional OpenClaw plugin bridge.
tags:
- openclaw
- gateway
- watchdog
- resilience
- wechat
- wsl
- systemd
- macos
- windows
requirements:
tools:
- bash
- curl
- systemctl optional
- launchctl optional
- powershell optional
- openclaw recommended
permissions:
- Writes a user-level systemd service, macOS LaunchAgent, or Windows scheduled task when requested.
- Writes config and logs under user-level config/state directories.
- Starts a localhost dashboard on 127.0.0.1:18790 when Python is available.
- Restarts OpenClaw Gateway through systemctl --user or openclaw gateway restart.
- Optional model probe sends real OpenClaw model requests only when explicitly enabled.
---

# OpenClaw Gateway Resilience Guard

Use this skill when a user wants to keep OpenClaw Gateway and message channels online after network drops, WSL sleep/resume, macOS/Windows wake events, long-lived connection failures, or recurring model-provider timeouts.

## Install

Linux, WSL, or macOS:

```bash
bash install-watchdog.sh
```

The installer works with defaults. It prompts for the main channel probe URL when running interactively, but pressing Enter is enough for the default WeChat probe.

For unattended install:

```bash
bash install-watchdog.sh --yes
```

For a custom channel:

```bash
bash install-watchdog.sh --channel-url "https://your-channel.example.com/health"
```

Windows PowerShell:

```powershell
powershell -ExecutionPolicy Bypass -File .\install-watchdog.ps1
```

After install, open the standalone dashboard:

```text
http://127.0.0.1:18790/
```

The dashboard is served by the watchdog process, not Gateway, so it remains the recovery entry when Gateway is down.

Optional Gateway-side bridge while Gateway is healthy:

```bash
openclaw plugins install ./openclaw-plugin
openclaw plugins enable resilience-guard
openclaw gateway restart
```

## Operate

```bash
systemctl --user status gateway-watchdog
journalctl --user -u gateway-watchdog -f
bash ~/.local/share/openclaw-gateway-watchdog/uninstall-watchdog.sh
```

On macOS, inspect `launchctl print gui/$(id -u)/ai.clawhub.gateway-resilience-guard`.
On Windows, inspect `Get-ScheduledTask -TaskName "OpenClaw Gateway Resilience Guard"`.

If user systemd is unavailable, the installer starts a direct background fallback and stores its pid under `~/.local/state/openclaw-gateway-watchdog/watchdog.pid`.

## Safety Model

The watchdog restarts only after layered checks:

1. OpenClaw native health probes: `openclaw

README.md

# OpenClaw Gateway Resilience Guard

External recovery guard for OpenClaw Gateway and long-lived channel plugins such as `openclaw-weixin`.

This project is for people who run OpenClaw continuously on Linux, WSL, macOS, or Windows and need the gateway to recover from channel disconnects, network sleep/resume, and long-lived session failures without babysitting the terminal.

## Problem

OpenClaw Gateway can still be alive while an individual channel is no longer healthy. This is common after laptop sleep, Wi-Fi changes, WSL network hiccups, or long idle periods.

The WeChat plugin is especially sensitive because it depends on a long-poll `getUpdates` loop. In the upstream `Tencent/openclaw-weixin` code, the monitor has a limited retry loop and session guard:

- `monitor.ts` defines `MAX_CONSECUTIVE_FAILURES = 3` and `BACKOFF_DELAY_MS = 30_000`.
- `session-guard.ts` defines `SESSION_PAUSE_DURATION_MS = 60 * 60 * 1000` and `SESSION_EXPIRED_ERRCODE = -14`.
- Issue [Tencent/openclaw-weixin#141](https://github.com/Tencent/openclaw-weixin/issues/141) reports that after a config hot reload the monitor can end without starting again; the workaround is `openclaw gateway restart`.
- Issue [Tencent/openclaw-weixin#155](https://github.com/Tencent/openclaw-weixin/issues/155) reports that `errcode=-14` can enter a 60-minute session pause loop and block outbound messages.

This watchdog does not replace the official plugin. It is an external safety net: when the gateway or channel stops behaving like a live system, it restarts the gateway with guardrails.

## Design

The script uses a layered health model before it restarts anything:

| Layer | Probe | Purpose |
| --- | --- | --- |
| Gateway | `openclaw gateway status --json --require-rpc`, local health URL, local TCP port, service/process fallback | Detect whether OpenClaw Gateway is down or locally unreachable. |
| Channel | `openclaw health --json --verbose`, `openclaw status --deep`, optional `openclaw channels status --probe`, then URL fallback | Prefer OpenClaw's own per-channel health model, then fall back to a configured URL when native probes are unavailable. |
| Runtime diagnostics | `openclaw models status --json`, `openclaw logs --plain`, warning classification | Separate provider, proxy/network, auth/rate-limit, gateway, channel-session, config reload, and task-runtime evidence before choosing an action. |
| Model API, optional | `openclaw agent --json` with the configured model provider | Detect whether the configured model path is timing out while Gateway and channels still look healthy. Disabled by default because it makes real model calls. |
| Network | multiple independent URLs, default Baidu/QQ/Weixin | Avoid restarting the gateway during whole-machine or ISP network failure. |

Only gateway failures restart immediately. Channel failures go through confirmation, network split-brain protection, exponential backoff, and restart-rate limits.
Model probe failures default to evidence logging only;

_meta.json

{
  "ownerId": "kn70e6hn9pv8yrpzjn5kqczygd8708vw",
  "slug": "gateway-resilience-guard",
  "version": "1.4.4",
  "publishedAt": 1779383203702
}

CHANGELOG.md

# Changelog

## 1.4.4

- Replace fragile Web Component controls with semantic native buttons and selects styled by the local dashboard design system.
- Fix the top action bar, sidebar navigation, refresh/export controls, and strategy buttons so they always render as clickable UI even if component scripts are blocked or stale.
- Center the brand mark text and restore pointer cursor, hover, focus, and active states across navigation and action controls.
- Fix config summary row layout so long environment keys no longer overlap values.

## 1.4.3

- Rebuild the dashboard visual system on local Fluent UI Web Components instead of plain native controls.
- Replace the old dropdowns and buttons with componentized controls, dark glass panels, animated ambient background, and refined hover states.
- Rebalance every dashboard page so Overview, Trends, Strategy, Logs, and Config each have complete work areas rather than sparse split-up cards.
- Add richer summaries for trend digest, latest signals, strategy matrix, action center, and runtime map.
- Serve bundled local component assets under `/vendor/` so the recovery UI does not depend on CDN availability.

## 1.4.2

- Rework the dashboard sidebar into real view switching for Overview, Trends, Strategy, Logs, and Config instead of one long page.
- Tighten the dashboard grid, panel sizing, button alignment, and sidebar polish for a cleaner operations-console layout.
- Fix the trend chart axis label collision by removing the redundant x-axis title and reserving more space for tick labels.
- Keep charts stable across refreshes by drawing only the active view and ignoring hidden canvases.
- Move strategy explanations into centered button content and remove the sidebar localhost note.

## 1.4.1

- Redesign the dashboard with a sidebar layout, stronger visual grouping, and selectable Light, Dark, Ocean, and Forest themes.
- Add Chinese/English language switching for dashboard labels, buttons, hints, legends, and notifications.
- Fix canvas redraw sizing so repeated refreshes do not stretch dashboard panels.
- Replace "overnight timeline" with a general event trend chart that includes lanes, legends, and time-axis ticks.
- Auto-follow the newest watchdog log lines while keeping a toggle for manual scrolling.
- Add an unlock/lock action flow and strategy hover hints so guarded controls are discoverable.

## 1.4.0

- Add a standalone local dashboard at `http://127.0.0.1:18790` that remains available when OpenClaw Gateway is down.
- Add dashboard API summaries, log-signal charts, overnight timelines, status file freshness, safe config summaries, diagnostic export, and quick strategy presets.
- Add guarded dashboard actions for run diagnostics, restart Gateway, and apply presets using a generated local action token.
- Add an OpenClaw native plugin bridge that registers `/resilience-guard` and redirects to the external dashboard when Gateway is healthy.
- Install dashboard files on Linux/WSL, macOS, and Windows.

## 1.3

README-watchdog.md

# OpenClaw Gateway 看门狗

这是项目的中文快捷入口。完整中文文档见 [README.zh-CN.md](README.zh-CN.md)，英文文档见 [README.md](README.md)。

最简单安装：

```bash
bash install-watchdog.sh
```

Windows：

```powershell
powershell -ExecutionPolicy Bypass -File .\install-watchdog.ps1
```

默认配置即可启动；需要指定自己的通道地址时：

```bash
bash install-watchdog.sh --channel-url "https://你的通道地址/health"
```

安装后还有一个独立图形化控制台：

```text
http://127.0.0.1:18790/
```

它由 watchdog 自己提供，不依赖 OpenClaw Gateway。Gateway 挂掉时，这个页面仍然可以看状态、图表、策略、日志和诊断导出。Dashboard 支持中英文切换、主题切换、事件趋势图、异常分类图和受保护操作解锁。

安装后它会默认采集 OpenClaw 诊断快照和日志信号，包括 gateway status、health、models status、status --deep 和 `openclaw logs --plain` 的 WARN/ERROR 摘要。关键证据在：

```text
~/.local/state/openclaw-gateway-watchdog/watchdog.log
~/.local/state/openclaw-gateway-watchdog/last-openclaw-diagnostics.jsonl
~/.local/state/openclaw-gateway-watchdog/last-openclaw-log-signals.txt
```

可选模型探针默认关闭。需要排查模型 API 是否在某个时段超时时，可以在配置里显式开启 `MODEL_PROBE_ENABLED="1"`；它会发送极小的 `openclaw agent --json` 请求，并把结果写入 watchdog 日志和 `model-probe-history.jsonl`。默认动作仍是只记录证据，不会擅自改 OpenClaw 配置。

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.