boost

Refines rough feature ideas into structured implementation prompts with requirements and success criteria.

Updated Jun 13, 2026
One-click install
npx skills add https://github.com/jenreh/project-kit-template --skill boost-jenreh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: boost
Source: https://github.com/jenreh/project-kit-template/tree/main/.agents/agent-skills/skills/boost
Command: npx skills add https://github.com/jenreh/project-kit-template --skill boost-jenreh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Vague feature requests like "add auth" or "send email notifications" lead to ambiguous implementations, scope creep, and wasted iterations. This Skill turns rough ideas into precise, actionable implementation prompts before any code is written. ## Core Features & Use Cases - Context-Aware Exploration: Reads project files, memory, and framework docs to ground the refined prompt in your actual codebase and existing patterns. - Targeted Clarification: Asks only 2-5 focused questions that cannot be answered from the code, avoiding redundant back-and-forth. - Structured Prompt Output: Produces a plan with Task, Context, Requirements, Deliverables, Success Criteria, and Out of Scope sections, delivered through plan-mode approval for iteration. - Use Case: You type "/boost I want to add a settings page to my Reflex app" and receive a detailed spec covering state management, routing, persistence, and verification steps that another AI session can execute directly. ## Quick Start Ask the assistant to boost your rough feature idea, for example by saying "boost this: add email notifications when a task is completed".

Frequently Asked Questions about boost

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

FAQPage Schema
How do I turn a vague feature idea into a detailed implementation prompt?

Invoke the boost workflow with your rough idea, such as "/boost add a settings page". It explores your codebase, asks a few targeted clarifying questions, then produces a structured prompt with requirements, deliverables, and success criteria.

What should a good AI coding prompt include?

A strong implementation prompt includes a task summary, relevant codebase context, numbered verifiable requirements, expected deliverables, success criteria with verification commands, and an explicit out-of-scope section to prevent scope creep.

Does the boost skill write implementation code?

No. The skill explicitly never writes implementation code and enforces this through plan mode. Its only output is a refined prompt that another session or developer can execute without follow-up questions.

When should I not use prompt refinement before coding?

Skip refinement for direct implementation requests, debugging questions, code reviews, or architectural discussions where you are not asking for a scoped prompt. It is designed for vague ideas that need scoping, not well-defined tasks.

How does prompt refinement use my existing project context?

It reads key files like CLAUDE.md and README.md, searches related code and patterns, queries stored memory for past decisions, and fetches current framework documentation. This ensures the refined prompt references your actual conventions and libraries.