oma-pm

Decomposes product requirements into prioritized task plans with dependencies and acceptance criteria.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/javiruu/viru-air --skill oma-pm-javiruu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: oma-pm
Source: https://github.com/javiruu/viru-air/tree/main/.agents/skills/oma-pm
Command: npx skills add https://github.com/javiruu/viru-air --skill oma-pm-javiruu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning ambiguous or complex product requests into actionable, dependency-aware implementation plans is slow and error-prone. This Skill structures requirements into prioritized tasks with clear ownership, acceptance criteria, and API contracts so specialist agents can execute in parallel. ## Core Features & Use Cases - Task Decomposition: Breaks feature requests into single-agent tasks with priority tiers, dependencies, complexity estimates, and testable acceptance criteria. - API-First Planning: Defines API contracts and data models before implementation tasks to prevent tight coupling and circular dependencies. - Standards-Aligned Planning: Optionally applies ISO 21500 project structure, ISO 31000 risk framing, and ISO 38500 governance guidance for enterprise deliveries. - Use Case: Given a request like "Build a TODO app with JWT authentication", it produces a JSON plan assigning backend, frontend, and QA tasks with dependencies, saved to .agents/results/plan-{sessionId}.json for orchestrator execution. ## Quick Start Ask the agent to plan and decompose your feature request into prioritized tasks with acceptance criteria and dependencies.

Frequently Asked Questions about oma-pm

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

FAQPage Schema
How do I break down a feature request into tasks for multiple agents?▼

Provide the product goal, constraints, and acceptance expectations. The Skill decomposes the request into single-agent tasks, each with an assigned agent, title, priority tier, dependencies, and testable acceptance criteria, saved as a JSON plan for orchestrator execution.

What output format does the planning agent produce?▼

It produces a JSON plan following the task template schema, including tasks, API contracts, data models, and non-functional requirements. Artifacts are saved to .agents/results/plan-{sessionId}.json and a task-board-compatible markdown summary.

Does the PM agent write implementation code?▼

No, it only plans and decomposes work. Implementation is delegated to specialist agents such as backend, frontend, mobile, and QA, which consume the generated plan and API contracts.

How are circular dependencies between tasks resolved?▼

Circular dependencies are broken by defining a shared API contract or interface first as a priority-0 task with no dependencies. The conflicting tasks then depend on the contract rather than on each other.

When should ISO 21500, 31000, or 38500 planning be applied?▼

Apply these standards for enterprise or regulated projects with material risk, many stakeholders, or unclear decision rights. Avoid them for small features or bugs where formal framing adds ceremony without improving delivery clarity.

What happens when requirements are too ambiguous to plan?▼

The Skill lists explicit assumptions, clarifies uncertainty triggers such as business logic or security concerns, and escalates questions before decomposition. It never creates tasks while requirements remain ambiguous.