AI Agent Hub
Back to skills
Universal Product Manager icon

Universal Product Manager

Business Operations Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Follow the instructions at https://skillhub.cn/install/skillhub.md to install @user_8c7f0f65/productmanagerskills.

About this skill

Problem

Product work often gets stuck in two places. Vague requests turn directly into PRDs, user stories, or roadmaps, which can become template filling without decisions. Cross-domain work across discovery, positioning, delivery, metrics, and AI products needs a shared standard, otherwise assumptions stay unlabeled, outcomes stay unmeasurable, and roles stay generic.

How It Works

The skill routes the request to the matching knowledge area: discovery and research, strategy and positioning, artifacts and delivery, finance and metrics, career and leadership, or AI product craft. It then loads the relevant knowledge/ sections and templates/ files, fills them with concrete context, and applies quality gates: label [assumption], define measurable outcomes, specify roles and tradeoffs, and flag antipatterns such as metrics theater, feature factory, and solution smuggling.

Boundaries

Use it when you need PM-style probing, first drafts, decomposition, and delivery structure. It depends on the provided knowledge modules and templates. If the request is outside PM scope, redirect instead of forcing a framework. Complex requests can use guided, info-dump, or best-guess modes; simple requests can be answered directly.

Use Cases

  • Before kickoff, turn scattered customer interview notes into a problem statement, an opportunity-solution tree, and testable experiment hypotheses while flagging assumptions and success metrics.
  • Prepare a PRD review by drafting user stories, acceptance criteria, epic hypotheses, and a list of assumptions that still need validation.
  • Run a SaaS metrics review using MRR, NRR, CAC, and LTV to decide feature ROI and channel investment priority for the next quarter.
  • Design an AI feature validation plan using PoL probing, context engineering, and agent orchestration to define test goals and failure risks.

Best For

  • Engineers who need to turn vague requirements into PRDs, user stories, and acceptance criteria before review, with tradeoffs and assumptions labeled.
  • Product owners who must explain MRR, NRR, unit economics, and channel ROI before leadership reviews and decide investment priority for the next quarter.
  • Algorithm or platform engineers who want to move AI features from demos to validated workflows with clear test goals and failure modes.
  • Product managers preparing to move into director or VP roles and need help framing altitude, scope, and readiness for transition interviews.