risk-surfacing

Write actionable product risks as scenario, indicator, and response for PRDs and roadmaps.

3|2|Updated Aug 24, 2026
One-click install
npx skills add https://github.com/hero-engine/hero --skill risk-surfacing-hero-engine
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: risk-surfacing
Source: https://github.com/hero-engine/hero/tree/main/domains/pm/skills/risk-surfacing
Command: npx skills add https://github.com/hero-engine/hero --skill risk-surfacing-hero-engine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Most Risks sections in PRDs and roadmaps are generic worry lists ("technical complexity", "timeline risk") that no team can plan against. This Skill turns vague unease into decision-useful risks by enforcing a three-part structure: a concrete scenario, an observable early indicator, and a pre-decided response. ## Core Features & Use Cases - Three-part risk template: Every risk names the specific failure scenario, the early signal that it is starting, and the planned response with rough cost and timing. - Premortem procedure: A four-step method (assume failure, enumerate causes independently, convert causes to three-part risks, rank by likelihood and impact) for generating risks that optimism hides. - Risk vs. assumption routing: Distinguishes known risks from untested assumptions, routing the latter to assumption testing instead of writing speculative responses. - Use Case: When writing a PRD for a bulk-import feature, run a premortem to surface that users may abandon after one bad import, then record the indicator (first-import error rate) and response (add a dry-run preview) in the Risks section. ## Quick Start Ask the AI to write the Risks section of your PRD using the scenario, indicator, and response format, or to run a premortem on a committed roadmap bet.

Frequently Asked Questions about risk-surfacing

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

FAQPage Schema
How do I write a useful Risks section for a PRD?▼

Write each risk in three parts: a concrete scenario describing the specific failure, an observable early indicator that fires before the failure, and a pre-decided response with rough cost and timing. Avoid generic entries like "timeline risk" that no one can plan against.

How to run a premortem for a product bet?▼

Assume the bet has already failed, have each stakeholder independently write why it failed, convert each cause into a scenario-indicator-response risk, then rank by likelihood and impact and keep only the vital few. Independent-first enumeration avoids anchoring on the loudest voice.

What is the difference between a risk and an assumption?▼

A risk is a failure scenario you can already picture, so you plan an indicator and response. An assumption is a belief you are unsure holds; it belongs in an assumption test, not a Risks section, because writing a speculative response for an unpictured scenario is false confidence.

Why is "monitor it" not an acceptable risk response?▼

Monitoring alone names no action, so the team cannot decide whether to pre-invest or accept the risk. A valid response states the concrete action and its rough cost, such as sharding a job scoped at three days, ready to start when the indicator fires.

When should I not use likelihood-impact heat maps for risks?▼

Heat maps are decoration when the cells do not name concrete scenarios. Score a specific scenario with its indicator and response, not a category like "adoption risk", otherwise the ranking carries no planning value.