auto

Orchestrate task lifecycle stages within a single Claude session.

1|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/huacheng/moonview --skill auto-huacheng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto
Source: https://github.com/huacheng/moonview/tree/main/plugins/task-ai/skills/auto
Command: npx skills add https://github.com/huacheng/moonview --skill auto-huacheng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, python3, and includes scripts (resource) components.

What problem does it solve?

Coordinates the entire task lifecycle within a single Claude session, maintaining context and progress across planning, verification, execution, and merging.

Core Features & Use Cases

  • Single-session orchestration of plan, verify, check, exec, merge, highlight, and report stages.
  • Self-correcting loop with optional delegation to subagents and built-in stall and quota handling.
  • Context management via structured compaction prompts and recovery metadata to ensure resilience.

Quick Start

Enter /task-ai:auto <notebook_name> in the prompt to start the autonomous execution loop.

Frequently Asked Questions about auto

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

FAQPage Schema
How do I orchestrate a full task lifecycle in a single Claude session?

You can orchestrate a full task lifecycle in a single Claude session by running an autonomous execution loop that coordinates planning, verification, execution, and merging stages while maintaining continuous context.

How does context management work during complex task automation?

Context management during task automation uses structured compaction prompts and recovery metadata to preserve continuity, enabling context-aware decision making and self-correction across the entire execution process.

Do I need bash and python3 to run autonomous task execution loops?

Yes, you need bash and python3 installed as dependencies to run the autonomous task execution loop, as the orchestration scripts rely on these environments to execute and manage the lifecycle stages.

What is the best way to handle stalls during multi-stage task orchestration?

The best way to handle stalls during multi-stage task orchestration is to use a self-correcting loop with built-in stall detection and quota handling safeguards, ensuring reliable progress tracking and automatic recovery.

Can I delegate subtasks to subagents during end-to-end progress tracking?

Yes, you can delegate subtasks to subagents during end-to-end progress tracking, because the orchestration loop supports optional delegation alongside its self-correcting lifecycle management and verification stages.

Why does task orchestration lose context between planning and merging stages?

Task orchestration loses context between planning and merging stages when sessions are fragmented, but running a single-session lifecycle with structured recovery metadata prevents this and maintains continuous progress.