extract-criteria

Decompose specs into workstreams with criteria and output JSON.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/patrob/skills --skill extract-criteria
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: extract-criteria
Source: https://github.com/patrob/skills/tree/main/plugins/claude-skills/skills/extract-criteria
Command: npx skills add https://github.com/patrob/skills --skill extract-criteria

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates spec parsing to a single subagent and returns a structured, machine-readable JSON containing workstreams, acceptance criteria, and dependencies to enable parallel planning and execution.

Core Features & Use Cases

  • Delegates spec decomposition to a single subagent to produce a complete plan of workstreams with scope, dependencies, and criteria.
  • Outputs a JSON document with {workstreams[], ac_map, per_workstream_criteria[]} that downstream skills can consume.
  • Accepts multiple input modalities (roadmaps / PRD Markdown, a plan.md from pw-deliberate, or inline text) and auto-detects the relevant spec when possible.
  • Ensures a strict contract between producer and consumer pairs by building an ac_map and explicit data shapes.

Quick Start

Provide the spec path (or inline text) to the agent and request a JSON decomposition of workstreams, criteria, and dependencies.

Frequently Asked Questions about extract-criteria

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

FAQPage Schema
How do I decompose a PRD into actionable workstreams?

You can decompose a PRD into actionable workstreams by feeding the roadmap Markdown or inline text to the agent. It delegates spec parsing to a subagent and returns a structured JSON containing independent workstreams, acceptance criteria, and dependencies.

What is the JSON schema structure for spec decomposition outputs?

The JSON schema structure for spec decomposition outputs contains three core elements: workstreams[], ac_map, and per_workstream_criteria[]. This consistent schema ensures a strict contract between producer and consumer pairs.

Can I use inline text instead of a plan.md for workstream extraction?

Yes, you can use inline text instead of a plan.md for workstream extraction. The system accepts multiple input modalities including PRD Markdown, plan.md files from pw-deliberate, or inline descriptions, and can auto-detect the most recent spec.

Does spec decomposition support parallel planning and execution?

Spec decomposition supports parallel planning and execution by breaking down specifications into independent workstreams with explicit dependencies. The machine-readable JSON output enables downstream skills to consume the structured plan.

What is the best way to convert a spec into machine-readable JSON?

The best way to convert a spec into machine-readable JSON is to delegate the parsing to a single subagent. This ensures a complete plan of workstreams with scope, dependencies, and criteria is generated in a consistent, structured schema.

Why do I need an acceptance criteria map for workstream decomposition?

You need an acceptance criteria map (ac_map) for workstream decomposition to enforce a strict contract between producer and consumer pairs. It links workstreams to explicit data shapes and criteria, ensuring downstream processes have clear validation targets.