auto-legion

Orchestrate parallel agent waves with verification gates for stream plans.

6|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/Corvalis-LLC/Crow-Stack --skill auto-legion
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-legion
Source: https://github.com/Corvalis-LLC/Crow-Stack/tree/main/skills/auto-legion
Command: npx skills add https://github.com/Corvalis-LLC/Crow-Stack --skill auto-legion

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Legion-based orchestration enables coordinating multiple AI agents in waves, reducing context size per agent while preserving correctness and speeding up complex streams.

Core Features & Use Cases

  • Breaks plans into TDD waves (T/I/D/R) with focused prompts and minimal context per agent.
  • Dispatches background agents in parallel per wave and centralizes verification between waves.
  • Verifies outputs and assembles results to ensure end-to-end correctness across the stream.
  • Supports both manual (/stream) and dominion-style execution flows based on plan annotations.

Quick Start

Launch a legion-enabled stream by applying a legion-annotated plan to begin coordinated agent waves and verify between waves.

Frequently Asked Questions about auto-legion

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

FAQPage Schema
How do I orchestrate parallel agent execution for complex automation streams?

Parallel agent execution is orchestrated by dispatching background agents in waves, centralizing verification between waves, and assembling validated results to ensure end-to-end correctness across the stream.

What is a TDD wave-based approach for decomposing automation tasks?

A TDD wave-based approach breaks plans into Test, Implementation, Dependent, and Retry waves, dispatching focused prompts with minimal context per agent to speed up complex streams while preserving correctness.

How do I verify outputs across multiple parallel AI agents?

Outputs across parallel AI agents are verified using strict post-wave validation gates that centralize verification between dispatched waves, ensuring end-to-end correctness before assembling the final results.

Can I execute stream-based automation without manual intervention?

Stream-based automation supports both manual and dominion-style execution flows, allowing plans to be annotated for autonomous, coordinated agent waves with automatic verification between each step.

What are the limitations of using parallel agent waves for automation?

Parallel agent waves require tasks to be decomposable into tests, implementations, and dependents; if a stream cannot be broken into discrete waves with strict verification gates, this orchestration approach will not be effective.