process-decomposer

Decompose complex user requests into structured process entries with measurable outcomes.

3|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/dvy1987/agent-loom --skill process-decomposer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: process-decomposer
Source: https://github.com/dvy1987/agent-loom/tree/main/.agents/skills/process-decomposer
Command: npx skills add https://github.com/dvy1987/agent-loom --skill process-decomposer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes ambiguity and planning overhead by converting vague or complex requests into structured, outcome-driven process entries that can be replayed, reused, and owned by a project orchestrator.

Core Features & Use Cases

  • Registry-aware triage: checks existing process volumes for exact or partial matches and either reuses, adapts, or creates new process entries.
  • Complexity classification: decides between single-skill routing, problem-to-plan escalation, skill-chain decomposition, or agent-chain construction.
  • Outcome gating and decomposition: enforces measurable success criteria, decomposes tasks into steps with skill and tool references, flags knowledge gaps, and writes canonical process documentation.

Quick Start

Decompose this task into a measurable, step-by-step process: Add dark mode to the React app.

Frequently Asked Questions about process-decomposer

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

FAQPage Schema
How do I break down complex software tasks into step-by-step execution plans?

Complex software tasks are decomposed into structured, outcome-driven process entries by classifying complexity and breaking requests into steps with measurable success criteria, skill references, and tool assignments.

What is process decomposition in multi-step agent workflows?

Process decomposition converts vague requests into reusable process entries for agent orchestration, classifying them as single-skill routing, skill-chain decomposition, or agent-chain construction based on complexity.

Can I reuse existing task plans instead of creating new ones for similar projects?

Yes, the workflow checks a process registry for exact or partial matches to existing process volumes, then either reuses, adapts, or creates new process entries to avoid planning overhead.

How do I ensure my task triage steps have measurable success criteria?

Outcome gating enforces measurable success criteria during task triage, decomposing tasks into steps with explicit skill and tool references while flagging knowledge gaps in canonical documentation.

What's the best way to assign skills and tools to multi-step project plans?

Task decomposition calls skill-finder and tool-finder to assign appropriate references to each step, routing tasks through single-skill execution, skill-chains, or agent-chains based on complexity classification.

When should I escalate a task to a full problem-to-plan workflow?

Problem-to-plan escalation is triggered when complexity classification determines a task exceeds single-skill routing, requiring structured decomposition into multi-step execution plans with documented process entries.