agent-stuck-recovery

Community

Unstick AI agents, ship real work fast.

Authorfreedomw1987
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill fixes the common issue where a main AI agent loses all forward motion mid-task without timing out or having its session interrupted: tool calls are running, but no actual work is being shipped, leaving the user waiting and progress stalled.

Core Features & Use Cases

  • Early stuck detection: Identifies leading signals like stacked context compaction headers and unproductive search loops before full stuck state occurs.
  • Hard budget enforcement: Caps recovery at 4-5 tool calls to force shipping of real minimal artifacts instead of endless explanation or rework.
  • Use case: If your AI agent has run 10+ tool calls on a simple task with no files written, or is responding only with long explanations and no action, use this Skill to immediately ship a minimal spec, ADR, or retrospective and restore progress.

Quick Start

Instruct the stuck main agent to activate the agent-stuck-recovery protocol to ship a real minimal artifact within 4 to 5 tool calls.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: agent-stuck-recovery
Download link: https://github.com/freedomw1987/tree_monstor/archive/main.zip#agent-stuck-recovery

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 536,000+ vetted skills library on demand.