think

Analyze session audit logs to detect execution loops and adjust strategies.

Updated Jun 22, 2026
One-click install
npx skills add https://github.com/alangeb/tau --skill think-alangeb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: think
Source: https://github.com/alangeb/tau/tree/main/src/skills/think
Command: npx skills add https://github.com/alangeb/tau --skill think-alangeb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill resolves cognitive bottlenecks where an AI agent becomes stuck in repetitive loops, loses track of complex multi-part tasks, or needs to reassess its strategy when assumptions change mid-execution.

Core Features & Use Cases

  • Loop Detection: Automatically identifies repeating tool call patterns in audit logs to prevent infinite execution cycles.
  • Strategic Reassessment: Provides a dedicated reasoning pass to re-evaluate plans when unexpected results occur or task requirements shift.
  • Use Case: If an agent is repeatedly failing to execute a file operation due to a permission error, invoking this skill allows the agent to step back, analyze the audit log, and formulate a new approach rather than continuing to retry the same failing command.

Quick Start

Use the think skill to analyze the current session and determine why the agent is stuck in a loop.

Frequently Asked Questions about think

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

FAQPage Schema
How do I debug an AI agent stuck in a repetitive tool call loop?

Loop detection identifies repeating tool call patterns in session audit logs to prevent infinite execution cycles. It enables deep reasoning to automatically analyze agent history and break repetitive agentic workflows.

What is the best way to reassess an agentic workflow when task requirements shift mid-execution?

Strategic reassessment spawns a dedicated reasoning fork for state evaluation, performing meta-analysis to re-evaluate plans when unexpected results occur or task requirements shift mid-execution in agentic workflows.

How does deep reasoning help with complex planning failures in autonomous agents?

Deep reasoning resolves complex planning failures by performing systematic debugging of tool call sequences and meta-analysis. It resolves cognitive bottlenecks where an agent loses track of multi-part tasks by evaluating state and adjusting strategy.

Do I need session audit logs to perform agentic loop detection and strategy adjustment?

Yes, session audit logs are required. Access to these logs is necessary to analyze repeating tool call patterns, evaluate state, and perform the systematic debugging required for effective loop detection and strategy adjustment.

Why does my AI agent repeatedly retry the same failing command during execution?

Your agent repeatedly retries the same failing command due to cognitive bottlenecks. Invoking deep reasoning allows the agent to step back, analyze audit logs, and formulate a new approach rather than continuing to retry the failing tool call.