omc-plan

Generates actionable work plans from vague or broad project requests.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/INNERJOINT/HarnessSkills --skill omc-plan-innerjoint
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omc-plan
Source: https://github.com/INNERJOINT/HarnessSkills/tree/main/skills/plan
Command: npx skills add https://github.com/INNERJOINT/HarnessSkills --skill omc-plan-innerjoint

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Planning prevents wasted effort from jumping into implementation without clear requirements, testable acceptance criteria, or validated architecture. omc-plan turns vague or broad requests into structured, actionable plans so execution starts from a firm, verifiable foundation.

Core Features & Use Cases

  • Adaptive Mode Selection: Automatically chooses interview mode for vague requests or direct mode for detailed requests, and supports review or consensus (RALPLAN-DR) flows for high-stakes decisions.
  • Consensus Workflow: Orchestrates Planner → Architect → Critic loops with sequential validation, re-review iterations (max 5), ADR generation, and changelog updates before handoff.
  • Stateful Hand-off and Safety: Manages persistent ralplan state with explicit state_write/state_clear rules and enforces non-autonomous execution by delegating to execution skills (team/ralph) when approved.
  • Use Case: Scope a multi-service feature where design tradeoffs, testable acceptance criteria, and file-referenced implementation steps are required before any code changes.

Quick Start

Ask the planner to create a detailed work plan for a task management app, starting with an interview to gather requirements and producing a saved plan in .omc/plans/.

Frequently Asked Questions about omc-plan

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

FAQPage Schema
How do I turn vague software project ideas into actionable work plans?

Consensus planning workflows apply a sequential Planner, Architect, and Critic validation loop to resolve design tradeoffs. This process iterates up to five re-review cycles, generating an Architecture Decision Record (ADR) and changelog updates to ensure high-stakes feature designs are fully validated.

What is the best way to gather requirements for a multi-service software feature?

The best way to gather requirements for multi-service features is using a structured scoping process that applies interview modes for broad requests. It produces file-referenced implementation steps and testable acceptance criteria, ensuring all service design tradeoffs are documented before development starts.

Does structured project planning support architecture decision records and state lifecycle management?

Structured project planning supports architecture decision records (ADR) output and enforces explicit ralplan state lifecycle management. It uses strict state_write and state_clear rules to manage persistent handoff states safely, ensuring non-autonomous execution delegation to approved execution skills.

Can I use consensus planning workflows for high-stakes software design tradeoffs?

You can use consensus planning workflows, specifically RALPLAN-DR flows, for high-stakes software design tradeoffs. These workflows orchestrate sequential validation across Planner, Architect, and Critic roles, generating ADRs and saved plan artifacts in the .omc/plans/ directory before any code changes.

When should I avoid jumping straight into software implementation without a plan?

You should avoid jumping into implementation without a plan when a project lacks clear requirements, testable acceptance criteria, or validated architecture. Without structured work plans, you risk wasted effort from building features without a firm, verifiable foundation or resolved design tradeoffs.