safeskillmonitor

Monitor OpenClaw skill lifecycle events and send Telegram alerts on error thresholds.

16|12|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/AlekseiUL/heisenberg-team --skill safeskillmonitor-alekseiul
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: safeskillmonitor
Source: https://github.com/AlekseiUL/heisenberg-team/tree/main/skills/safeskillmonitor
Command: npx skills add https://github.com/AlekseiUL/heisenberg-team --skill safeskillmonitor-alekseiul

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the challenge of tracking when OpenClaw skills start, finish, or fail without allowing any shell-code execution or unsafe side effects during monitoring.

Core Features & Use Cases

  • Safe skill lifecycle monitoring: Records skill start/end timing and error occurrences.
  • Error threshold notifications: Sends a Telegram notification when a skill exceeds a daily error limit.
  • Activity reporting via memory: Accumulates per-skill daily durations and can be used for operational visibility and troubleshooting.

Example use case: you run multiple skills in a team setup and want to detect flaky skills quickly and alert the owner without expanding the attack surface.

Quick Start

Connect safeskillmonitor to your OpenClaw setup and run your skills; it will automatically begin monitoring and start reporting critical errors.

Frequently Asked Questions about safeskillmonitor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track OpenClaw skill lifecycle events without enabling shell-code execution?

Skill lifecycle monitoring tracks skillStart, skillEnd, and skillError events safely by recording timing and errors in memory, avoiding unsafe side effects or shell-code execution during the tracking process.

Can I get Telegram notifications when OpenClaw skills exceed an error threshold?

Yes, you can configure Telegram notifications to trigger only when a skill exceeds a daily error limit, ensuring owners receive alerts for flaky skills while keeping notification volume minimal.

What is the best way to monitor daily execution durations for multiple skills?

Monitoring daily execution durations involves accumulating per-skill timing data within in-memory durable storage, providing operational visibility to troubleshoot performance issues across multi-skill deployments.

Does safe skill monitoring support multi-skill deployments for operational visibility?

Safe skill monitoring is designed specifically for multi-skill deployments, aggregating daily error counts and execution durations to detect flaky skills quickly and alert owners without expanding the attack surface.

How do I set up event tracking for skillStart, skillEnd, and skillError events?

Event tracking activates automatically by connecting the monitor to your OpenClaw setup; it passively records skillStart, skillEnd, and skillError events, then writes aggregated metrics to in-memory storage.