OpenClaw Task Scheduler

Task queue management with WebSocket real-time updates. Schedule, monitor, and control background tasks. Supports immediate, scheduled, and recurring tasks w...

What to check before installing or running this skill: - The documentation and architecture list modules (websocket.py, persistence.py, task.py, queue.py, examples/) that are not present in the package; ask the publisher for the missing files or inspect the full scheduler.py for the missing functionality. - Review the remainder of lib/scheduler.py (the scan truncated save_state and possibly other functions) to confirm there are no unexpected network endpoints, hardcoded remote hosts, telemetry, or code that reads arbitrary files or environment variables. - Confirm how persistence works: SKILL.md mentions tasks.db and persistence.py, but the CLI as provided constructs an in-memory TaskScheduler for each command — test whether starting the daemon actually persists tasks and where files are written. - The SKILL.md lists 'exec' among allowed tools and shows running the daemon and using wscat; running the daemon will open local ports (default ws://localhost:8080). Only run in a safe environment if you don't want a local service listening on ports or writing files. - Inspect requirements (websockets, python-crontab, aiosqlite) and install them from trusted sources, and run the daemon in an isolated environment (container or VM) first to observe behavior. If you want, I can (1) show the remainder of the truncated scheduler.py for further review, (2) search the codebase for any network calls or file-write APIs, or (3) produce a short checklist to run the daemon safely in a sandbox.

Type: OpenClaw Skill Name: oc-task-scheduler Version: 1.0.0 The skill bundle provides a standard task scheduling framework with a CLI and library components for managing background tasks and real-time updates via WebSockets. While the SKILL.md requests high-privilege tools (exec, cron), the provided Python code in lib/scheduler.py and scripts/main.py contains only legitimate task management logic, async worker loops, and placeholders for persistence, with no evidence of malicious intent, data exfiltration, or harmful prompt injection.