rcode-herdr-orchestration

Orchestrate multiple Claude agents across isolated git worktrees with TSC-gated merges.

2|1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/hanzlahabib/rcode --skill rcode-herdr-orchestration
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rcode-herdr-orchestration
Source: https://github.com/hanzlahabib/rcode/tree/main/rcode/skills/actions/4-implementation/rcode-herdr-orchestration
Command: npx skills add https://github.com/hanzlahabib/rcode --skill rcode-herdr-orchestration

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill coordinates multiple coding agents on one repository so you can fan out research, audits, and fixes without losing control of branches, worktrees, or merge order.

Core Features & Use Cases

  • Parallel agent orchestration: Split work across isolated worktrees and monitor each agent through herdr panes.
  • Campaign-based delivery: Manage multi-wave efforts with durable backlog, state tracking, heartbeat checks, and an integration branch.
  • Safe merge discipline: Keep work local, gate merges with type-checking, and avoid pushing or landing changes without explicit approval.
  • Use case: A large refactor or audit can be divided into several focused sub-tasks, executed in parallel, then merged back in a controlled sequence.

Quick Start

Ask the assistant to launch a herdr-based parallel campaign for this repository, split the work into isolated agents, and track the backlog and merges until the task is complete.

Frequently Asked Questions about rcode-herdr-orchestration

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

FAQPage Schema
How do I run parallel AI coding agents on the same repository safely?

You can run parallel AI coding agents safely by orchestrating them across isolated git worktrees, monitoring each through dedicated panes, and merging their outputs into a campaign integration branch with type-checking gates.

What is git worktree orchestration for multi-agent engineering tasks?

Git worktree orchestration for multi-agent tasks is a process that assigns each AI agent its own isolated working directory to execute parallel audits or fixes, preventing branch conflicts before merging.

How do I manage multi-wave fix campaigns with multiple AI agents?

You manage multi-wave fix campaigns by tracking a durable backlog and state files, using heartbeat checks to monitor agent progress, and sequencing merges back into the integration branch until completion.

Can I use TypeScript type-checking to gate agent merges in a parallel coding workflow?

Yes, you can use TypeScript type-checking to gate merges in a parallel coding workflow, ensuring that code from isolated agents passes TSC validation before integrating into the campaign branch.

How do I prevent accidental pushes when merging output from parallel AI agents?

You prevent accidental pushes by enforcing explicit no-push safeguards during the merge process, keeping all agent output local until you explicitly approve landing the changes.

What are the limitations of using isolated worktrees for parallel agent orchestration?

Limitations include the requirement for dedicated pane control, one worktree per agent, committed backlog files, and strict adherence to no-push safeguards to avoid uncontrolled branch integration.