run-autopilot

Automate the end-to-end PRD lifecycle across multiple PRDs with deterministic phase transitions.

2|Updated Dec 20, 2020
One-click install
npx skills add https://github.com/buvis/home --skill run-autopilot
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: run-autopilot
Source: https://github.com/buvis/home/tree/main/.claude/skills/run-autopilot
Command: npx skills add https://github.com/buvis/home --skill run-autopilot

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Orchestrates the end-to-end PRD lifecycle automatically, reducing manual overhead and enabling faster iteration by coordinating catchup, planning, work, review, rework, blind review, doubt review, and completion across multiple PRDs.

Core Features & Use Cases

  • Deterministic phase orchestration: catchup, planning, work, review, rework, blind review, doubt review, and done.
  • Stateful execution: tracks PRD progress, per-phase status, and batch progress in a local state store.
  • Autonomous decision logging: records auto-fix decisions, research-backed outcomes, and user-deferrable items for batch-end review.
  • Batch management and reporting: aggregates per-PRD outcomes into batch reports and deferred logs for governance.
  • Safe operation with pausing: triggers safety pauses for ambiguity or blockers and supports external stop-hook integration.

Quick Start

Invoke /run-autopilot to begin the end-to-end PRD lifecycle automation for your current backlog.

Frequently Asked Questions about run-autopilot

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

FAQPage Schema
How do I automate the entire PRD lifecycle across multiple product requirements?

Automating the PRD lifecycle across multiple requirements is done by orchestrating phases like catchup, planning, work, review, and rework autonomously. The Skill coordinates these deterministic transitions while tracking progress in a local state store.

Can I manage state and track progress for product development workflows locally?

You can manage state locally by tracking per-PRD progress, phase status, and batch progress in a local state store. This stateful execution supports dev, local, and autopilot environments to maintain deterministic workflow transitions.

How does autonomous decision logging work for automated project management?

Autonomous decision logging records auto-fix decisions, research-backed outcomes, and user-deferrable items during execution. These records are aggregated into batch reports and deferred logs for batch-end governance and review.

What happens when an automated PRD workflow encounters ambiguity or blockers?

When ambiguity or blockers occur during PRD execution, the workflow triggers safety pauses and supports external stop-hook integration. This prevents autonomous operations from proceeding without necessary user input or resolution.

Does automated PRD orchestration support blind review and doubt review phases?

Automated PRD orchestration supports blind review and doubt review phases as part of its deterministic lifecycle. These phases are integrated alongside catchup, planning, work, rework, and completion to ensure comprehensive product requirement validation.

What is the best way to handle batch-end reporting for multiple PRDs?

Handling batch-end reporting for multiple PRDs is managed by aggregating per-PRD outcomes into batch reports and deferred logs. This provides comprehensive governance over autonomous decisions made throughout the execution lifecycle.