skills.rest
myysophia
myysophiaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

autonomous-skill

Orchestrate long-running autonomous tasks across Codex sessions with Initializer and Executor roles.

197|29|Updated Oct 21, 2025
Search Tags:#workflow#codex#autonomy#ai-agent#task-management#multisession
One-click install
npx skills add https://github.com/myysophia/codex-config --skill autonomous-skill-myysophia
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autonomous-skill
Source: https://github.com/myysophia/codex-config/tree/main/skills/autonomous-skill
Command: npx skills add https://github.com/myysophia/codex-config --skill autonomous-skill-myysophia

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automates and orchestrates long-running tasks that require multiple Codex sessions, tracking progress and resuming work across Initializer and Executor roles in a non-interactive mode.

Core Features & Use Cases

  • Automatic task decomposition with persistent task state in a dedicated .autonomous directory.
  • Non-interactive Codex execution with auto-continuation and session resumption via stored session IDs.
  • Isolated task work streams for parallel autonomous tasks and safe failure recovery.

Quick Start

Run the run-session.sh script with a task description to start a new autonomous task.

Frequently Asked Questions about autonomous-skill

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

FAQPage Schema
How do I coordinate autonomous tasks across multiple Codex sessions?▼

You can coordinate autonomous tasks by running the run-session.sh script with a task description, which triggers an Initializer and Executor pattern to decompose tasks and track progress in a local .autonomous directory.

What is the best way to automate long-running Codex workflows with minimal manual intervention?▼

Automating long-running Codex workflows requires non-interactive execution with auto-continuation, which automatically resumes tasks using stored session IDs and persistent task state for safe failure recovery.

Do I need a specific environment to run multi-session autonomous AI agent tasks?▼

Yes, running multi-session autonomous AI agent tasks requires a Codex CLI environment, as the orchestration depends on Codex sessions and stores persistent task state under a dedicated .autonomous directory.

How does task state recovery work when an autonomous Codex session fails?▼

Task state recovery works by storing session IDs and decomposed task state in a .autonomous directory, allowing the system to safely resume isolated task work streams after a failure.

Can I run parallel autonomous tasks in Codex without them interfering with each other?▼

Yes, you can run parallel autonomous tasks using isolated task work streams, which maintain separate state within the .autonomous directory to prevent interference and ensure safe failure recovery.

When should I not use an autonomous task orchestration pattern for AI agents?▼

You should not use autonomous task orchestration for short-lived or interactive tasks, as this pattern is designed specifically for long-running, non-interactive multi-session workflows requiring auto-continuation and session resumption.

Related Skills

Explore More Software Engineering Skills→
mathruffian-dotmathruffian-dot
70

codex-env-setup

Set up Codex environments across desktop, CLI, and IDE extensions.

Community
Intermediate
fukukei23fukukei23

resume-session

Load recent handoff records and reconstruct active work state.

Community
Advanced
justinsenseijustinsensei
1

kanban-workflows

Coordinate Kanban-based multi-agent workflows with task decomposition and scheduling.

Community
Advanced
smykla-skalskismykla-skalski

harness:session:start

Coordinate multi-agent orchestration sessions with JSON outputs for automation.

Official
Advanced