self-diagnosis

Diagnose AI system health by monitoring metrics and detecting anomalies.

12|1|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/blackplume233/Actant --skill self-diagnosis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-diagnosis
Source: https://github.com/blackplume233/Actant/tree/main/examples/actant-hub/skills/self-diagnosis
Command: npx skills add https://github.com/blackplume233/Actant --skill self-diagnosis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Monitors AI agent ecosystems to detect health issues early, identify anomalies, and locate root causes before they cause outages.

Core Features & Use Cases

  • Metric collection across Daemon, Agent, and MCP components (CPU, memory, disk, latency)
  • Anomaly-pattern detection (spike, drift, cascade, recurrence) to trigger proactive alerts
  • Dependency-chain tracing to map failures from Daemon to external services
  • Log analysis for error extraction and hotspot identification
  • Use Case: In production, automatically surface root causes for high-latency tasks and failed agents

Quick Start

Run a health-check pass across active agents and report any detected anomalies.

Frequently Asked Questions about self-diagnosis

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

FAQPage Schema
How do I monitor AI agent ecosystems to detect health issues early?

To monitor AI agent ecosystems, collect metrics across Daemon and Agent components to detect anomalies and locate root causes before outages. This involves tracking CPU, memory, disk, and latency to trigger proactive alerts.

What is the best way to trace root causes for high-latency tasks in production agents?

The best way to trace root causes for high-latency tasks is dependency-chain tracing, which maps failures from Daemon components to external services. This localizes faults across logs and MCP traces.

Can I use anomaly detection for proactive alerting on AI system health?

Yes, anomaly detection enables proactive alerting on AI system health by identifying patterns like spikes, drifts, cascades, and recurrences. This ensures early detection of performance degradation in production agents.

Does log analysis work for error extraction and hotspot identification in MCP traces?

Log analysis works for error extraction and hotspot identification by examining MCP traces and system logs. It helps surface specific failure points and map them across the agent dependency chain.

How do I run a health-check pass across active agents to report anomalies?

Run a health-check pass by applying guided diagnostics across active agents to collect system metrics and analyze logs. This reports detected anomalies and pinpoints the root causes of failures.