automation-decomposition

Convert implementation plans into dependency-resolved Automation Issue Manifests.

1|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools --skill automation-decomposition
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: automation-decomposition
Source: https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools/tree/main/.cursor/skills/automation-decomposition
Command: npx skills add https://github.com/Atlas-Memory-Framework/AtlasMemory-Tools --skill automation-decomposition

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It eliminates the manual, error-prone work of translating an accepted implementation plan into a precise, machine-readable contract for issue projection and unattended issue-to-PR automation.

Core Features & Use Cases

  • Automation Issue Manifest generation: Converts implementation plan sections into a downstream-ready ## Automation Issue Manifest that defines containers, leaf issues, dependencies, and gates.
  • Dispatch-mode aware decomposition: Supports none, manifest-only, issue-projection, and unattended-prs with rules that determine what metadata must be present.
  • Risk-to-execution boundaries: Applies policy so high-risk areas default to manual-review or blocked, while eligible work is bounded for agent execution.

Quick Start

Ask the automation decomposition skill to draft the exact ## Automation Issue Manifest for your approved implementation plan, targeting AutomationTarget=unattended-prs.

Frequently Asked Questions about automation-decomposition

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

FAQPage Schema
How do I convert an implementation plan into a dependency graph for unattended PRs?

An automation manifest converts an implementation plan into bounded, dispatchable leaf issues with dependency resolution and validation gates for unattended issue-to-PR automation.

What is an automation manifest and when do I need one for issue projection?

An automation manifest projects an accepted plan into containers, leaf issues, and gates. You need one when decomposing plans for issue projection or unattended issue-to-PR automation.

How do I handle dispatch modes when decomposing plans for GitHub automation?

Dispatch-mode handling requires specifying metadata based on the target mode: none, manifest-only, issue-projection, or unattended-prs, ensuring correct boundaries before agents execute work.

Can I set high-risk areas to manual review before generating an automation manifest?

Yes, risk dispatch policy applies risk-to-execution boundaries so high-risk areas default to manual-review or blocked, while eligible work is bounded for unattended agent execution.

What validation gates are required for unattended issue-to-PR automation?

Gate-backed validation requires strict dependency and gate enforcement before agents execute work, ensuring the single-block manifest draft is valid for downstream projection tooling.