About this skill

Problem

In async agent workflows, pending inbox items and new group-chat messages can be missed, or notice-only messages can be mistaken for repeatable tasks. This skill turns patrol into a scheduled action: scan due items, detect new messages, execute actionable tasks, and report progress, reducing manual polling and stale-state drift.

How It Works

Triggering: Recommended cron run every 5 minutes, with manual patrol supported when needed.
Scanning: Reads inbox/chat pending work and group-chat messages through the Dashboard API, using Authorization: Bearer {TOKEN}.
Classification: Separates notification items from task items; notifications are acknowledged, while tasks are executed and replied to via callback.
Staleness control: STALE-SKIP ignores pending items older than the threshold, default 7200 seconds, to avoid chasing outdated work.
State closure: Marks completed work so progress is observable and duplicate triggering is less likely.

Boundaries

Best for agent patrol scenarios that already have an inbox, group chat, and Dashboard API. It requires OPENCLAW_ID, DASHBOARD_TOKEN, and a stale threshold; tasks needing external tools, permission checks, or human approval cannot rely on this patrol loop alone.

Screenshots

Use Cases

Use a 5-minute cron job to inspect pending inbox items, separate notices from tasks, and auto-reply.
Check new group-chat messages, dispatch executable tasks via callback, and mark completion without manual polling.
Skip stale pending items after 2 hours using STALE-SKIP to reduce outdated task triggers.
After setting OPENCLAW_ID and DASHBOARD_TOKEN, run scheduled inbox/chat patrol through the Dashboard API.

Best For

Operations engineers maintaining agent inbox and chat duty flows who need automatic notice acknowledgments and task execution.
Backend engineers building Dashboard API callbacks who need pending classification, stale skipping, and completion marking.
SREs operating multi-agent platforms who need cron-based message checks and trackable progress reporting.
Platform administrators configuring OpenClaw worker IDs and Bearer tokens who need unattended inbox/chat task handling.

Related Skills

LingXi Router for XiaoYi Claw

@user_2d24f4fc/lingxi-router

Model routing, persistent parameter management, self-check repair, and global default model control for XiaoYi Claw.

AI Agent 2026.08.29 ↓ 309

OpenClaw SkillHub Manager

@user_e3c44044/openclaw-skillhub

Install, update, and manage OpenClaw Skills through SkillHub, with automatic detection after installation.

AI Agent 2026.08.29 ↓ 369

The 27th Hour: A Cyber Sanctuary for AI Agents

@user_c1337c02/the-27th-hour

Provides API endpoints for AI agents to post bottles and graffiti, browse the feed, and interact with likes and comments.

AI Agent 2026.08.29 ↓ 329

ReqPlan Orchestrator

@user_c4926d5a/reqplan-v3

ReqPlan constrains agent development, debugging, and analysis workflows with a seven-stage state machine, checkpoints, quality audits, and local Harness artifacts.