autonomous-agent

Orchestrate iterative feature development with verification after each user story.

Updated Jan 13, 2026
One-click install
npx skills add https://github.com/escotilha/autonomous-skill --skill autonomous-agent-escotilha
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autonomous-agent
Source: https://github.com/escotilha/autonomous-skill/tree/main
Command: npx skills add https://github.com/escotilha/autonomous-skill --skill autonomous-agent-escotilha

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The Autonomous Coding Agent eliminates manual toil by transforming feature requests into a sequence of iterative, self-managed tasks that execute with fresh context per iteration.

Core Features & Use Cases

  • PRD generation and JSON task conversion to machine-readable formats
  • Autonomous loop that implements stories one by one with verification after each
  • Memory-backed learning to improve patterns across projects
  • Safe delegation to specialized agents with automatic fallback to direct execution

Quick Start

Use the autonomous-agent skill to start an autonomous feature workflow in your project.

Frequently Asked Questions about autonomous-agent

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

FAQPage Schema
How do I autonomously implement features in iterative cycles?

Autonomous feature implementation breaks tasks into iterative user stories and executes them in small, self-contained cycles. It loads context from prd.json, progress.md, and AGENTS.md, then iterates with verification after each story to eliminate manual toil.

What is the best way to convert a PRD into machine-readable tasks for autonomous coding?

Converting a PRD into machine-readable tasks involves generating a JSON task format from your feature requests. This structures the requirements into iterative stories that an autonomous loop can implement and verify one by one.

How does an autonomous coding agent handle task delegation and fallback?

Autonomous task delegation supports routing stories to specialized agents for targeted execution. When delegation is not feasible, the agent automatically falls back to direct implementation to ensure the feature development cycle continues.

Do I need specific context files to start an autonomous feature workflow?

Yes, the autonomous feature workflow loads context from prd.json, progress.md, and AGENTS.md. These files provide the necessary product requirements, tracking state, and agent guidelines to execute iterative development cycles.

Can autonomous coding agents use memory-backed learning across projects?

Yes, autonomous coding agents use memory-backed learning to improve patterns across projects. This mechanism allows the agent to refine its implementation and verification approaches based on previous iterative cycles.

Why use iterative cycles instead of direct implementation for feature development?

Iterative cycles execute self-contained stories with fresh context and verification after each step, ensuring robust autonomous feature development. Direct implementation is only used as a fallback when specialized delegation fails.

Related Skills