shepherd

Orchestrates a multi-agent sprint pipeline with gated phases and automated quality reviews.

Updated May 12, 2026
One-click install
npx skills add https://github.com/FL03/claude-shepherd --skill shepherd-fl03
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: shepherd
Source: https://github.com/FL03/claude-shepherd/tree/main/skills/shepherd
Command: npx skills add https://github.com/FL03/claude-shepherd --skill shepherd-fl03

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill solves the failure modes of long-running engineering sessions, such as tunnel vision, code duplication, and process drift, by enforcing a structured, audited sprint pipeline.

Core Features & Use Cases

  • Closed-Flock Execution: Orchestrates a six-agent team (Engineer, Critic, Coder, Auditor, Worker, Discovery) with defined roles and mechanical guardrails.
  • Sprint-Execution Contract: Enforces repeatable, audited pipelines with gated phases, per-project memory, and automated drift detection.
  • Use Case: Use this when initiating a multi-hour engineering sprint to ensure that all work is planned, reviewed, and audited against a strict set of architectural and quality standards.

Quick Start

Invoke the shepherd skill to spawn a new sprint execution session for the current branch.

Frequently Asked Questions about shepherd

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

FAQPage Schema
How do I prevent process drift during long-running engineering sprints?

To prevent process drift during long-running engineering sprints, use an orchestrated multi-agent pipeline that enforces gated phases, per-project memory, and automated drift detection to ensure all work remains audited against strict architectural standards.

What is multi-agent sprint pipeline automation?

Multi-agent sprint pipeline automation orchestrates a defined team of agents like Engineer, Critic, and Auditor to execute software development lifecycles, managing parallel lane execution and enforcing git custody to produce production-grade, audit-ready output.

How do I execute parallel lanes in a sprint pipeline?

To execute parallel lanes in a sprint pipeline, invoke a closed-flock orchestration session for your current branch, coordinating parallel lane execution while enforcing strict architectural invariants and branch lifecycle management to prevent drift.

Can I use automated quality reviews for branch lifecycle management?

Yes, automated quality reviews are integrated directly into branch lifecycle management, enforcing git custody and applying automated reviews within gated phases to ensure all branch output meets strict architectural invariants before merging.

When should I use a closed-flock execution model for software engineering?

Use a closed-flock execution model when initiating a multi-hour engineering sprint, solving failure modes like tunnel vision and code duplication by enforcing a structured, audited pipeline with mechanical guardrails and defined agent roles.

Best way to audit software development lifecycles for architectural invariants?

The best way to audit software development lifecycles for architectural invariants is using a gated sprint pipeline with an Auditor agent, enforcing strict architectural compliance and automated drift detection throughout the branch lifecycle.