demo-verify

Builds minimal runnable prototypes to validate design risks before full development.

2|Updated Jun 8, 2026
One-click install
npx skills add https://github.com/HACK-WU/skills --skill demo-verify-hack-wu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: demo-verify
Source: https://github.com/HACK-WU/skills/tree/main/skills/demo-verify
Command: npx skills add https://github.com/HACK-WU/skills --skill demo-verify-hack-wu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Before committing to full development, teams often discover too late that a design relies on unproven technology, fails performance targets, or breaks on third-party integrations. This Skill validates those risk points with minimal throwaway prototypes so design flaws surface early and cheaply. ## Core Features & Use Cases - Risk-Driven Validation: Automatically identifies technical feasibility, performance, integration, and alternative-scheme risks from a design document, then builds the smallest possible runnable prototype for each. - Clear Verdicts: Every risk point receives an explicit conclusion (pass / conditional pass / fail) with evidence, driving a continue / adjust / rollback decision. - Requirement Integration: When a .requirements/config exists, verification reports and demo code are persisted into the requirement directory via the req CLI. - Use Case: A design proposes Redis distributed locks and a 1000 QPS query endpoint. The Skill builds small prototypes in .demo-verify/, benchmarks them, and reports that the lock works but the query needs a cache layer before development proceeds. ## Quick Start Ask the AI to verify the risky parts of the current design document with a quick demo prototype before starting full development.

Frequently Asked Questions about demo-verify

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

FAQPage Schema
How do I validate a technical design before writing production code?

Identify the uncertain risk points in the design document, then build the smallest runnable prototype that exercises each one. This Skill automates that flow: risk identification, minimal prototype in a .demo-verify/ directory, execution, and an explicit pass/fail verdict per risk.

When should I build a proof-of-concept prototype versus coding directly?

Build a prototype when the design uses unfamiliar libraries or APIs, involves performance-sensitive paths, integrates third-party services, or has multiple competing schemes. Skip it for simple CRUD work, pure UI changes, or when a similar proven implementation already exists.

What should a design verification prototype include?

Only the minimum needed to reach a conclusion on the risk point. Prototypes should be self-contained, runnable, and disposable, skipping error handling, logging, and configuration. Each file should state its verification purpose in a header comment.

What happens if the prototype verification fails?

A failed risk point triggers a rollback to the design phase rather than forced development. First debug to distinguish a genuinely infeasible approach from a bug in the prototype itself; fixable prototype bugs can be corrected and re-verified.

Does the verification prototype get kept after development?

Prototypes live in a temporary .demo-verify/ directory during development as reference material and are deleted after development completes. If requirement management is configured, demo code is migrated into the requirement's demo/ directory for permanent archival.