Long Run

Automates end-to-end backlog processing from selection to completion.

1|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/beekeeper-lab/foundry --skill long-run
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Long Run
Source: https://github.com/beekeeper-lab/foundry/tree/main/.claude/skills/long-run
Command: npx skills add https://github.com/beekeeper-lab/foundry --skill long-run

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Puts the Team Lead into autonomous backlog processing mode. The Team Lead reads the bean backlog, selects the best bean to work on, decomposes it into tasks, executes the full team wave, verifies acceptance criteria, commits the result, and loops to the next bean. Continues until no actionable beans remain or an unrecoverable error occurs.

Core Features & Use Cases

  • Orchestrates end-to-end backlog processing, from backlog assessment to completion.
  • Decomposes backlog beans into task sequences for multiple roles (BA, Architect, Developer, Tech-QA).
  • Supports sequential and parallel execution with optional tmux-backed parallelism, and manages branches and commits.

Quick Start

Start a long-run session to drive autonomous backlog processing across the bean pipeline until all actionable beans are completed.

Frequently Asked Questions about Long Run

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

FAQPage Schema
How do I automate backlog processing for my project management workflow?

Automate backlog processing by guiding the Team Lead to select, decompose, and close beans sequentially. The workflow orchestrates end-to-end backlog assessment, execution, verification, and closure until no actionable beans remain.

How does task decomposition work for multi-role development waves?

Task decomposition splits backlog beans into task sequences for BA, Architect, Developer, and Tech-QA roles. The workflow supports both sequential and parallel execution with optional tmux-backed parallelism to drive the full team wave.

How do I enforce git branching and acceptance criteria verification during autonomous execution?

Autonomous execution enforces git branching and acceptance criteria verification during the execution and verification phases. It manages branches and commits automatically before closing the bean and looping to the next item.

Does autonomous backlog processing work with parallel execution environments?

Autonomous backlog processing supports parallel execution using optional tmux-backed parallelism. It manages multi-role waves including BA, Architect, Developer, and Tech-QA within both sequential and parallel execution modes.

What happens when an unrecoverable error occurs during automated team wave execution?

Automated team wave execution stops processing when an unrecoverable error occurs or no actionable beans remain. The workflow loops continuously through mode detection, bean selection, execution, and closure phases otherwise.

Can I use Long Run to merge branches after completing backlog beans?

Long Run supports optional merging after completing backlog beans. The workflow implements closure phases with status updates and acceptance criteria verification before proceeding to the optional merge step.