avoid-feature-creep

Enforces scope discipline and feature validation during product planning and AI coding sessions.

1|Updated Oct 10, 2025
One-click install
npx skills add https://github.com/Rocktown-Labs/rivercitymd --skill avoid-feature-creep-rocktown-labs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: avoid-feature-creep
Source: https://github.com/Rocktown-Labs/rivercitymd/tree/main/.cursor/skills/avoid-feature-creep
Command: npx skills add https://github.com/Rocktown-Labs/rivercitymd --skill avoid-feature-creep-rocktown-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Feature creep delays launches, bloats codebases, and burns teams out. This Skill gives developers and AI agents a structured framework to validate feature requests, defend MVP scope, and say no to unvalidated additions before they derail a release. ## Core Features & Use Cases - Feature Decision Framework: A five-step checklist covering problem validation, alignment, impact measurement, complexity assessment, and a final gut check before any feature is approved. - Scope Management Rules: MVP scope document templates, version-controlled scope tracking, the 48-hour rule for new requests, and budget-based trade-off negotiation. - AI Session Discipline: Session start, mid-session, and end-of-day checks to prevent AI coding agents (Claude, Cursor, Codex) from expanding scope through unsolicited suggestions. - Use Case: While building an MVP with an AI assistant, the agent proposes adding caching, refactors, and extra error handling. Use this Skill to log each suggestion in a scope decision log, defer non-essential items, and keep the session focused on the defined core feature. ## Quick Start Ask the AI to review your current feature list against the decision framework and flag anything that does not solve the validated core user problem.

Frequently Asked Questions about avoid-feature-creep

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

FAQPage Schema
How do I prevent feature creep in my MVP?▼

Define a written MVP scope document listing what is in scope, explicitly out of scope, and your success criteria before coding starts. Apply the five-step decision checklist to every new request and defer anything that lacks validated user evidence.

How do I stop AI coding agents from adding extra features?▼

State your scope constraints at the start of every AI session and treat agent suggestions like stakeholder requests. Log suggestions in a scope decision log, apply the 48-hour rule, and run mid-session checks asking whether scope has expanded.

What is the 48-hour rule for feature requests?▼

The 48-hour rule means waiting two days before adding any new feature request to the backlog. Most requests that feel urgent lose their urgency after reflection, filtering out impulsive scope additions.

When should I say no to a stakeholder feature request?▼

Say no when the request lacks validated user evidence, does not align with the core product vision, or would delay the current release. Offer a trade-off instead: explain what would be cut or delayed to make room for it.

How do I fix a product that already has feature bloat?▼

Audit every feature against usage data, categorize them as core, supporting, peripheral, or bloat, then deprecate features with under 5% engagement. Add feature flags and usage-metric requirements to prevent recurrence.