swarm

Orchestrate parallel SDLC workflows with isolated git worktrees and automated verification.

15|4|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/jmylchreest/aide --skill swarm-jmylchreest
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: swarm
Source: https://github.com/jmylchreest/aide/tree/main/skills/swarm
Command: npx skills add https://github.com/jmylchreest/aide --skill swarm-jmylchreest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates multiple agents to run parallel SDLC workflows with isolated worktrees, enabling scalable development pace and reliable collaboration.

Core Features & Use Cases

  • Launch multiple story agents, each with its own SDLC pipeline (DESIGN→TEST→DEV→VERIFY→DOCS).
  • Enforce per-story isolation via git worktrees and orchestrator-driven task management.
  • Handle VERIFY failures with a build-fix loop and seamless merging with /aide:worktree-resolve.

Quick Start

Spawn 3 story agents to run parallel SDLC pipelines across multiple stories.

Frequently Asked Questions about swarm

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

FAQPage Schema
How do I run parallel software development workflows on multiple independent features?

You can run parallel software development workflows by deploying multiple story agents that each operate in isolated git worktrees. This orchestration applies independent DESIGN, TEST, DEV, VERIFY, and DOCS stages to each feature.

What is the best way to isolate parallel feature branches during automated development?

Isolating parallel feature branches is handled by enforcing per-story isolation via git worktrees. The orchestrator deploys multiple story agents into these isolated worktrees to ensure reliable collaboration and scalable development.

How do automated verification loops handle build failures in orchestrated pipelines?

Automated verification loops handle build failures by applying a build-fix loop to resolve VERIFY stage failures. This loop uses seamless merging with worktree-resolve to ensure merge-ready completion of the story.

Can I spawn multiple story agents to run isolated SDLC pipelines simultaneously?

Yes, you can spawn multiple story agents to run isolated SDLC pipelines simultaneously. Each agent manages its own automated pipeline covering DESIGN, TEST, DEV, VERIFY, and DOCS stages for a specific story.

What are the limitations of orchestrating parallel worktrees for software development?

Orchestrating parallel worktrees is limited to scenarios with independent features requiring full SDLC pipelines. It is not suited for stories with overlapping dependencies or shared code modifications that break worktree isolation.

Does parallel SDLC orchestration require specific git worktree configurations?

Parallel SDLC orchestration requires git worktrees to enforce per-story isolation. The orchestrator manages task distribution across these worktrees to execute the automated verification loop and merge-ready completion.