using-agents-stack

Coordinate goal-QA pipeline phases with file-based state and status.json gates.

28|2|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/labs21-dev/agents-stack --skill using-agents-stack-labs21-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: using-agents-stack
Source: https://github.com/labs21-dev/agents-stack/tree/main/templates/.agents/skills/using-agents-stack
Command: npx skills add https://github.com/labs21-dev/agents-stack --skill using-agents-stack-labs21-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This root orchestrator coordinates the entire lifecycle of the goal-QA pipeline across multiple phases, ensuring problem statements, evidence, and state are durably stored and consistently acted upon by specialized workers.

Core Features & Use Cases

  • Orchestrates phase sequencing (goal -> spec -> plan -> tasks -> implement -> qa -> release) with blocking gates and status tracking.
  • Enforces adversarial separation (Generator ≠ Auditor) and provides deterministic dispatch of phase workers.
  • Maintains a file-backed state model and orchestrator-driven handoffs for auditable execution traces and reliable cold-start recovery.

Quick Start

Initialize a new workstream by creating goal.md and status.json, then trigger the initial goal phase.

Frequently Asked Questions about using-agents-stack

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

FAQPage Schema
How do I orchestrate an end-to-end agent workflow with phase gating?

Agent workflow orchestration coordinates the goal-QA pipeline across SPEC, PLAN, TASKS, IMPLEMENT, QA, and RELEASE phases using blocking gates and status tracking. It ensures deterministic routing using file-based state and status.json checks.

What is adversarial separation in multi-agent pipelines?

Adversarial separation enforces Generator and Auditor role isolation within agent pipelines. The orchestrator provides deterministic dispatch of phase workers to maintain separation of duties and ensure auditable execution traces across the workflow.

How to handle cold-starts and cross-phase handoffs in state-machine workflows?

State-machine workflow cold-starts and cross-phase handoffs are managed through a file-backed state model. The orchestrator-driven handoffs ensure reliable recovery and maintain traceability between durable artifacts stored on disk.

How do I initialize a new workstream for goal-QA driven orchestration?

Initialize a goal-QA workstream by creating a goal.md file and a status.json file. Triggering the initial goal phase starts the orchestrator, which then sequences the subsequent pipeline phases with blocking gates.

Does this agent orchestration stack support traceability between artifacts?

Yes, the agent orchestration stack satisfies traceability between artifacts by enforcing an auditable execution trace. It maintains durable storage of problem statements and evidence acted upon by specialized workers across phases.

When should I use a file-based state model for agent workflow orchestration?

Use a file-based state model for agent workflow orchestration when deterministic routing and durable state storage are required. It ensures reliable cold-start recovery and consistent phase gating without external database dependencies.