agent-introspection-debugging

Diagnose and recover from AI agent failures with structured introspection.

2|Updated Apr 14, 2026

Search Tags:#AI agent #systematic analysis #agent failure #self-debugging #recovery workflow

Ready to run

npx skills add https://github.com/klu-dev/porting-ecc-to-vscode --skill agent-introspection-debugging-klu-dev

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-introspection-debugging
Source: https://github.com/klu-dev/porting-ecc-to-vscode/tree/main/.github/skills/agent-introspection-debugging
Command: npx skills add https://github.com/klu-dev/porting-ecc-to-vscode --skill agent-introspection-debugging-klu-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a structured self-debugging workflow for AI agent failures, enabling systematic diagnosis and recovery.

Core Features & Use Cases

Structured Workflow: Guides agents through failure capture, root-cause diagnosis, contained recovery, and introspection reporting.
Use Case: When an AI agent is failing repeatedly or consuming tokens without progress, this Skill can help diagnose and recover from these failures.

Quick Start

Use the agent-introspection-debugging skill to capture and diagnose an agent failure.

Frequently Asked Questions about agent-introspection-debugging

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

FAQPage Schema

How do I debug an AI agent that keeps failing repeatedly?▼

To debug an AI agent failing repeatedly, you need a structured self-debugging workflow that captures the failure, diagnoses the root cause, executes a contained recovery, and generates an introspection report. This systematic approach stops repeated retries and token waste.

What is systematic self-debugging for AI agent failures?▼

Systematic self-debugging for AI agent failures is a structured process encompassing failure capture, root-cause diagnosis, contained recovery, and introspection reporting. It provides a methodical way to analyze and resolve agent errors instead of relying on uncontrolled retries.

When do I need to use a structured recovery workflow for AI agents?▼

You need a structured recovery workflow for AI agents when facing repeated retries, context overflow, or environmental mismatches. It is specifically required when the agent is consuming tokens without making progress, indicating a systematic failure that requires containment and diagnosis.

How do I diagnose an AI agent consuming tokens without progress?▼

To diagnose an AI agent consuming tokens without progress, run a systematic analysis that captures the failure state and identifies the root cause. The process isolates the environmental mismatch or context overflow triggering the loop and applies containment actions to recover.

What's the best way to capture and report AI agent failure states?▼

The best way to capture and report AI agent failure states is using a structured introspection workflow. It systematically captures the error, diagnoses the root cause, executes a contained recovery, and generates a detailed introspection report for analysis.

Does this self-debugging workflow handle context overflow and environmental mismatches?▼

Yes, this self-debugging workflow explicitly handles context overflow and environmental mismatches. It applies systematic analysis and containment actions to diagnose these specific agent failures, guiding the agent through a structured recovery process to restore normal operation.