Agent Introspection Debugging

by affaan-m

Agent Introspection Debugging is an AI Engineering skill for Claude Code, published by affaan-m in ECC.

239.8K stars36.4K forkson affaan-m/ECCAdded 2026/08/13Repository updated 2026/08/12

View source on GitHub Download repo ZIP

ai-agentsanthropicclaudeclaude-codedeveloper-toolsllmmcpproductivity

Install in seconds

Install Agent Introspection Debugging

Copy Agent Introspection Debugging into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/affaan-m/ECC/tree/main/skills/agent-introspection-debugging ~/.claude/skills/agent-introspection-debugging

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/affaan-m/ECC.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/agent-introspection-debugging/SKILL.md in affaan-m/ECC
Installs to: ~/.claude/skills/agent-introspection-debugging
Collection: One of 25 skills cataloged from this repository
Category: AI Engineering — 2451 skills

What Agent Introspection Debugging does

Agent Introspection Debugging guides an AI agent through a structured four-phase loop to capture failures, diagnose root causes, apply contained recovery actions, and produce an introspection report. Use it when an agent run fails repeatedly, loops, or drifts and you need a reproducible diagnosis rather than a blind retry.

Agent Introspection Debugging is cataloged under AI Engineering on DirSkills. Agent Introspection Debugging comes from a repository tagged ai-agents, anthropic, claude, claude-code and developer-tools.

Documentation

README

Excerpt

Agent Introspection Debugging

Use this skill when an agent run is failing repeatedly, consuming tokens without progress, looping on the same tools, or drifting away from the intended task.

This is a workflow skill, not a hidden runtime. It teaches the agent to debug itself systematically before escalating to a human.

When to Activate

Maximum tool call / loop-limit failures
Repeated retries with no forward progress
Context growth or prompt drift that starts degrading output quality
File-system or environment state mismatch between expectation and reality
Tool failures that are likely recoverable with diagnosis and a smaller corrective action

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Agent Introspection Debugging

What else does affaan-m publish alongside Agent Introspection Debugging?
Agent Introspection Debugging is one of 25 skills that DirSkills catalogs from affaan-m/ECC, the repository it ships in. Its siblings there include AI Regression Testing, AI-First Engineering and API Connector Builder. Each one is a separate skill with its own page in this directory, installs the same way Agent Introspection Debugging does, and is maintained by affaan-m in that same repository. The rest of the collection is listed on the affaan-m/ECC page.
How does Agent Introspection Debugging compare to other AI Engineering skills?
Agent Introspection Debugging ranks #5 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Agent Introspection Debugging against them. Open each page to compare what they document and how they install.