execute

Launch an autonomous TDD loop for stories in a JSON file.

Updated Feb 23, 2026
One-click install
npx skills add https://github.com/progremir/pluto --skill execute-progremir
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: execute
Source: https://github.com/progremir/pluto/tree/main/skills/execute
Command: npx skills add https://github.com/progremir/pluto --skill execute-progremir

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of writing and testing code by launching an autonomous TDD loop, ensuring stories are implemented efficiently and reliably.

Core Features & Use Cases

  • Autonomous TDD Loop: Spawns a fresh Claude instance for each story, implementing it using Test-Driven Development.
  • Self-Healing: Automatically attempts to fix failures and retries, escalating to blocking if issues persist after multiple attempts.
  • Phase-Boundary UAT: Pauses execution at the end of each phase for human review and approval.
  • Use Case: Kick off the core development cycle for a new feature, allowing the AI to handle the iterative process of writing tests, implementing code, and fixing bugs until all stories are complete.

Quick Start

Start the autonomous execution loop for all stories in stories.json.

Frequently Asked Questions about execute

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

FAQPage Schema
How do I automate test-driven development for multiple software stories?

You can automate test-driven development by launching an autonomous TDD execution loop that processes software stories defined in a JSON file. The loop spawns a fresh AI context for each story to iteratively write tests, implement code, and fix failures.

What is a self-healing code execution loop?

A self-healing code execution loop automatically attempts to fix implementation failures and retries within a fresh AI context. If issues persist after multiple configurable attempts, it escalates by blocking further execution.

How do I set up autonomous TDD execution for my development cycle?

To set up autonomous TDD execution, define your software development stories in a JSON file. You then launch the execution loop to process these stories, allowing the system to handle the iterative development cycle autonomously.

Does the autonomous TDD loop stop for human review?

Yes, the autonomous TDD loop features phase-boundary UAT, which pauses execution at the end of each phase. This safety mechanism ensures a human can review and approve the implemented stories before proceeding.

Can I configure the maximum retry attempts for self-healing code fixes?

Yes, you can configure the maximum iterations for the self-healing process. This limits the number of attempts the AI makes to fix failures before escalating the issue as blocking.

What are the limitations of autonomous TDD execution loops?

The main limitation of an autonomous TDD execution loop is that persistent failures will block progress after reaching the maximum iteration limit. It relies on phase-boundary UAT pauses to ensure human intervention for unresolved issues.