orchestration

Coordinate multi-issue GitHub work with parallel agents and dependency management.

Updated Jan 7, 2026
One-click install
npx skills add https://github.com/fyrsmithlabs/marketplace --skill orchestration-fyrsmithlabs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: orchestration
Source: https://github.com/fyrsmithlabs/marketplace/tree/main/plugins/contextd/skills/orchestration
Command: npx skills add https://github.com/fyrsmithlabs/marketplace --skill orchestration-fyrsmithlabs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables orchestration of multi-task work across GitHub issues and epics by coordinating parallel agents, managing dependencies, and enforcing consensus reviews. It leverages contextd for memory, checkpoints, and context folding to provide robust, auditable progress.

Core Features & Use Cases

  • Parallel task execution: Dispatch multiple issues or epics to agents concurrently while respecting dependencies.
  • Dependency and branching discipline: Resolve task graphs, create isolated feature branches, and prevent mainline edits during orchestration.
  • Consensus-driven quality: Run phased reviews after groups to ensure high quality before proceeding, with remediation and PR creation.
  • Contextd integration: Use memory, checkpoints, and context folding to preserve state, enable resume, and improve learning over time.
  • Use Case: Orchestrate an epic spanning several issues, automatically branch, run tasks in parallel, accumulate results, and open a PR when complete.

Quick Start

Trigger orchestration with the contextd command that fits your workflow, for example: /contextd:orchestrate --issues 42,43 --epic EPIC-1. The skill requires a contextd MCP server and will guide you through resolution, branching, execution, and review steps.

Frequently Asked Questions about orchestration

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

FAQPage Schema
How do I coordinate parallel agents to work on multiple GitHub issues concurrently?

You can coordinate parallel agents across multiple GitHub issues by dispatching tasks concurrently while managing dependencies. The orchestration process enforces isolated feature branches to prevent mainline edits during execution.

What is the best way to manage dependencies and branching when executing an epic across several issues?

The best way to manage dependencies when executing an epic is to resolve the task graph and create isolated feature branches for each issue. This branching discipline ensures traceability and prevents direct mainline edits during orchestration.

Do I need a contextd MCP server to orchestrate multi-issue work with checkpoints?

Yes, you need a contextd MCP server to orchestrate multi-issue work with checkpoints. The contextd integration provides memory retention, context folding, and checkpointing to preserve state and enable resume capabilities during execution.

How does consensus review work for parallel agent execution in software engineering?

Consensus review for parallel agent execution runs phased reviews after groups of tasks complete. This ensures high quality before proceeding, triggering automated remediation and PR creation when issues are detected.

Can I resume a multi-issue orchestration task if it stops partway through execution?

Yes, you can resume a multi-issue orchestration task if it stops. The contextd integration uses memory retention and checkpoints to preserve state, enabling robust resume, traceability, and recovery for incomplete phased executions.

When should I avoid running parallel agents for GitHub issue orchestration?

You should avoid running parallel agents for GitHub issue orchestration when tasks have strict sequential dependencies that cannot be resolved into a task graph, or when your workflow requires direct mainline edits instead of isolated feature branches.