autonomous-skill

Decompose user requests into multi-session tasks with persistent progress tracking.

1.6k|244|Updated Jul 18, 2025

Search Tags:#automation #project management #task management #multi-session #autonomous agent #long-running tasks #claude cli

Ready to run

npx skills add https://github.com/feiskyer/claude-code-settings --skill autonomous-skill

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: autonomous-skill
Source: https://github.com/feiskyer/claude-code-settings/tree/main/plugins/autonomous-skill/skills/autonomous-skill
Command: npx skills add https://github.com/feiskyer/claude-code-settings --skill autonomous-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires claude-cli, and includes scripts (resource) components.

What problem does it solve?

This Skill manages and executes complex, multi-session tasks autonomously, eliminating the need for constant human oversight and manual progress tracking, letting you focus on higher-level strategy.

Core Features & Use Cases

Dual-Agent Pattern: Utilizes an Initializer Agent to decompose tasks and an Executor Agent to complete them, ensuring structured progress.
Auto-Continuation: Automatically resumes tasks across sessions with persistent progress tracking via task_list.md and progress.md.
Task Isolation: Organizes each task in its own dedicated directory (.autonomous/<task-name>/) to prevent conflicts and maintain clarity.
Use Case: Use this skill to build a REST API for a todo app. It will break down the task, track progress, and continue across multiple sessions until completion.

Quick Start

Use the autonomous-skill to build a REST API for a todo app.

Frequently Asked Questions about autonomous-skill

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

FAQPage Schema

How do I automate long-running tasks across multiple sessions?▼

Automate long-running tasks by decomposing them into structured work units tracked across sessions. This Skill uses an Initializer Agent to break down your request and an Executor Agent to complete it, maintaining progress via task_list.md and progress.md so work resumes automatically between sessions.

Can I use autonomous agents to manage multi-session projects like API development?▼

Yes. This Skill applies a dual-agent pattern to multi-session projects including API development, code refactoring, and data processing. Each task runs in an isolated directory (.autonomous/<task-name>/), enabling structured decomposition, execution, and automatic continuation without manual intervention.

What's the best way to track progress on long-running development tasks?▼

Track progress using persistent task management with task_list.md and progress.md files. The Skill maintains per-task directories and automatically resumes work, eliminating manual progress tracking and letting you focus on strategy rather than session-to-session handoffs.

How does task isolation prevent conflicts in multi-session work?▼

Task isolation organizes each task in its own dedicated directory under .autonomous/<task-name>/. This prevents conflicts between concurrent or sequential tasks and maintains clarity by keeping initialization prompts, executor logic, and progress separate for each workload.

Do I need the Claude CLI to use this autonomous task management?▼

Yes. This Skill requires the Claude CLI and integrates with run-session.sh for seamless task orchestration. The Claude CLI enables multi-session agent execution and progress persistence across your autonomous workflows.

Can I automatically continue tasks without manual prompting between sessions?▼

Yes. Auto-continuation resumes tasks automatically across sessions using persistent progress tracking. The Executor Agent reads prior work from task_list.md and progress.md, then picks up where the previous session stopped without requiring manual intervention.