run

Automate end-to-end sprint execution with safety gates and human-in-the-loop review.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/project-purupuru/world-purupuru --skill run-project-purupuru
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: run
Source: https://github.com/project-purupuru/world-purupuru/tree/main/.claude/skills/run-mode
Command: npx skills add https://github.com/project-purupuru/world-purupuru --skill run-project-purupuru

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end sprint execution with safety gates, reducing manual overhead and risk of runaway automation.

Core Features & Use Cases

  • Autonomous sprint loop with guardrails, circuit breakers, and configurable limits
  • Draft PR creation and human-in-the-loop review workflow for every sprint
  • Red Team gate support and extensible sprint planning / rollback mechanisms

Quick Start

To begin, issue a run for a specific sprint (for example, /run sprint-1) and let the system manage guardrails and PR creation.

Frequently Asked Questions about run

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

FAQPage Schema
How do I automate sprint execution while maintaining safety guardrails?

Autonomous sprint execution with safety guardrails uses an automated run loop with configurable max cycles, timeouts, and human-in-the-loop review checkpoints to prevent runaway automation during development cycles.

What is a human-in-the-loop review workflow for draft PRs in autonomous sprints?

A human-in-the-loop review workflow for draft PRs requires the autonomous sprint loop to generate draft pull requests at decision points, pausing execution for manual review and approval before proceeding with further Git operations.

How do I start an autonomous run for a specific development sprint?

To start an autonomous run for a specific development sprint, issue a run command targeting your sprint identifier, such as /run sprint-1, and let the system manage guardrails, circuit breakers, and draft PR creation automatically.

Can I configure max cycles and timeouts for autonomous sprint loops?

Yes, autonomous sprint loops support configurable max cycles and timeouts, allowing teams to set hard limits on execution duration and iterations to ensure runaway automation triggers circuit breakers and stops safely.

Does autonomous sprint execution support Red Team gates and rollback mechanisms?

Autonomous sprint execution supports Red Team gate validation and extensible sprint planning rollback mechanisms, enabling teams to validate safety controls and revert sprint plans when guardrail-driven decision points detect issues.

When should I not use autonomous sprint execution for development cycles?

You should avoid autonomous sprint execution when development cycles lack clear guardrail criteria or human-in-the-loop review capacity, as the system relies on draft PR approvals and safety gates to mitigate the risk of runaway automation.