Agent Introspection Debugging

When your agent hits max tool calls, loops on the same command, or burns tokens without progress, this gives it a structured way to debug itself before you step in. It walks through capture, diagnosis, contained recovery, and a final introspection report. The core idea is solid: match the failure to a known pattern like context overflow or service timeouts, then take the smallest reversible action that changes the situation. The four phase loop keeps things systematic instead of retrying blindly with slightly different wording. Pairs naturally with verification-loop after code changes and works best when you need the agent to stop, assess what actually broke, and produce a legible report instead of just claiming it fixed something.

skills.sh installs: 5.7K
GitHub stars: 300.3K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add affaan-m/ecc --skill agent-introspection-debugging --agent claude-code

Installs into .claude/skills of the current project.

Files

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