prompt-enhancer-pack
Auto-enhances vague prompts into well-structured, context-aware prompts before Claude responds. Includes four skills: /enhance (general), /e
Open source Repository Open in the app JSON README (API)
About
Auto-enhances vague prompts into well-structured, context-aware prompts before Claude responds. Includes four skills: /enhance (general), /enhance-code (detects your tech stack from package.json, requirements.txt, etc. and tailors the prompt), /enhance-writing (classifies into 6 writing types and applies the right framework), and /enhance-for (rewrites for ChatGPT, Gemini, Cursor, or Copilot).
Details
- Kind
- Plugins
- Topic
- AI, RAG & memory
- Publisher
- sivajyothipamarthy
- Origin
- marketplace
- Category
- ferramentas
- Stars
- 1
- Last push
- 2026-04-22T12:13:53Z
- Repository state
- ativo
- Language
- Shell
- License
- MIT
- Added
- 2026-08-30 01:48:58
- Updated
- 2026-08-30 01:48:58
- Origin id
sivajyothipamarthy/prompt-enhancer-pack/prompt-enhancer-pack
README
# 🔧 Prompt Enhancer Pack for Claude Code
Stop writing vague prompts. This pack auto-enhances your prompts before Claude responds.
## Demo
## What's New in v2.0
- **Context-aware**: Scans your codebase before enhancing — detects tech stack, frameworks, and project patterns
- **Few-shot examples**: Each skill includes before/after examples so Claude knows exactly what "good" looks like
- **Named techniques**: Uses specific methods like specificity injection, anti-patterns, and constraint setting
- **Writing classification**: Automatically categorizes into 6 writing types and applies the right framework
- **Deep platform profiles**: Each AI platform gets genuinely different optimization with detailed examples
## Install (10 seconds)
**Recommended — install as a Claude Code plugin** (auto-updates, one command):
```
/plugin marketplace add SivaJyothiPamarthy/prompt-enhancer-pack
/plugin install prompt-enhancer-pack@prompt-enhancer
```
Skills are then available as `/prompt-enhancer-pack:enhance`, `/prompt-enhancer-pack:enhance-code`, etc.
To update later: `/plugin marketplace update prompt-enhancer`.
<details>
<summary>Manual install (legacy, unscoped skill names)</summary>
**macOS / Linux:**
```bash
git clone https://github.com/SivaJyothiPamarthy/prompt-enhancer-pack.git
cd prompt-enhancer-pack
chmod +x install.sh
./install.sh
```
**Windows:**
```cmd
git clone https://github.com/SivaJyothiPamarthy/prompt-enhancer-pack.git
cd prompt-enhancer-pack
install.bat
```
Then **restart Claude Code**. Skills are available as `/enhance`, `/enhance-code`, etc.
</details>
## Commands
> **Note on naming:** Examples below use the unscoped names (`/enhance`). If you installed via the plugin marketplace, the names are namespaced: `/prompt-enhancer-pack:enhance`, `/prompt-enhancer-pack:enhance-code`, etc. Functionality is identical.
| Command | Best For | Context-Aware | Example |
|---------|----------|:---:|---------|
| `/enhance` | Anything | ✅ | `/enhance explain kubernetes` |
| `/enhance-code` | Coding tasks | ✅ Detects tech stack | `/enhance-code add authentication` |
| `/enhance-writing` | Writing tasks | ✅ Matches content style | `/enhance-writing blog post about AI` |
| `/enhance-for` | Other AIs | — | `/enhance-for chatgpt explain docker` |
### `/enhance [prompt]` — General purpose
Analyzes your prompt for intent, audience, scope, and gaps. Adds specificity, structure, constraints, and anti-patterns.
```
/enhance explain kubernetes
/enhance compare SQL vs NoSQL
/enhance help me with system design interview prep
```
### `/enhance-code [prompt]` — Code focused (reads your codebase)
Automatically detects your project's tech stack from `package.json`, `requirements.txt`, `Cargo.toml`, etc. References actual files and patterns from your codebase in the enhanced prompt.
```
/enhance-code add authentication # detects Next.js, suggests NextAuth
/enhance-code add endpoint for users # detects FastAPI, matches existing router pattern
/enhance-code python web scraper # no project? falls back to best practices
```
**Example output inside a Next.js project:**
```
ENHANCED CODE PROMPT
─────────────────────────────────────
Project: Next.js 14 + TypeScript + Prisma
─────────────────────────────────────
Add authentication using NextAuth.js (already in package.json).
Implement: sign-in page at /auth/signin with email/password
and Google OAuth, session management using the existing Prisma
adapter, protected API routes using middleware...
```
### `/enhance-writing [prompt]` — Writing focused (classifies type)
Classifies your request into one of 6 categories (technical, business, marketing, creative, academic, personal) and applies the right enhancement framework for each.
```
/enhance-writing cover letter for Google # → Personal: authenticity, highlights
/enhance-writing blog post about AI trends # → Marketing: hook, audience, CTA
/enhance-writing API documentation # → Technical: structure, code examples
/enhance-writing email to client about delay # → Business: tone, action items
```
### `/enhance-for [platform] [prompt]` — Cross-platform
Each AI responds differently to prompt structures. This rewrites your prompt optimized for a specific platform's strengths. Copy the result and paste it there.
Supported: `chatgpt` · `gemini` · `cursor` · `copilot` · `claude`
```
/enhance-for chatgpt explain microservices
/enhance-for cursor build a login component in React
/enhance-for gemini compare cloud providers
```
| Platform | Optimization Strategy |
|----------|----------------------|
| ChatGPT | Role-based persona, markdown formatting, numbered steps |
| Gemini | Structured sections, tables, comparison frameworks |
| Cursor | File paths, function signatures, project-aware context |
| Copilot | Comment-style, input/output types, inline edge cases |
| Claude | Clear constraints, scope boundaries, anti-patterns |
## How It Works
```
You type: /enhance explain docker
↓
Claude: [Scans project context if available]
↓
Claude shows: Enhanced, well-structured prompt
↓
You choose: yes → proceed | edit → modify | skip → use original
↓
Claude: High-quality response based on approved prompt
```
## Project Structure
```
prompt-enhancer-pack/
├── README.md
├── CHANGELOG.md
├── LICENSE
├── .claude-plugin/
│ ├── marketplace.json → declares this repo as a Claude Code marketplace
│ └── plugin.json → plugin manifest (name, version, author)
├── install.sh → legacy manual installer
├── install.bat → legacy manual installer (Windows)
├── uninstall.sh → legacy manual uninstaller
├── assets/
│ └── demo.png
└── skills/
├── enhance/SKILL.md → /enhance
├── enhance-code/SKILL.md → /enhance-code
├── enhance-writing/SKILL.md → /enhance-writing
└── enhance-for/SKILL.md → /enhance-for
```
## Uninstall
**Plugin install:**
```
/plugin uninstall prompt-enhancer-pack@prompt-enhancer
/plugin marketplace remove prompt-enhancer
```
**Legacy manual install:**
```bash
chmod +x uninstall.sh
./uninstall.sh
```
## Contributing
PRs welcome! To add a new enhancer:
1. Create `skills/your-skill-name/SKILL.md`
2. Add YAML frontmatter with `name` and `description`
3. Add `allowed-tools` if it needs codebase access
4. Include few-shot examples for better results
5. Test it locally, then submit a PR
## Requirements
- Claude Code 2.x+