🧠 Prompt Library CLI

🚧 Work in Progress
This tool was recently refactored to use NestJS. Interactive menus are generally functional, but non-interactive commands need more testing and may be unstable. Comprehensive tests are still in progress. Use with caution and feel free to report any issues!

Prompt Library CLI is your personal, local-first toolkit for managing and executing sophisticated AI prompts. Store, organize, version control, and run prompts for Claude (Anthropic) and GPT (OpenAI) models directly from your terminal.

Treat AI prompts as reusable mini-applications that transform AI models into specialized agents.

📚 Table of Contents

🎯 Purpose & Features
- Key Features
⚡ Quick Start
🛠️ How It Works
- Recommended Workflow (CLI-First)
- Alternative Workflow (Git-Based)
🖥️ CLI Usage
- Main Commands
- Common Tasks
📂 Prompt Showcase
- Your AI Agent Collection
🚀 Getting Started
🧩 Prompt Composition with Fragments
- Using Fragments
⚙️ Metadata Generation
🤝 Contributing
📄 License

🎯 Purpose & Features

This CLI promotes a structured approach to prompt engineering:

Modularity: Build complex interactions using reusable prompt fragments.
Organization: Categorize prompts with rich, automatically generated metadata.
Version Control: Manage your prompt library using Git, locally or with a remote.
Reusability: Apply successful prompts across different projects and contexts.
Execution: Run prompts interactively or non-interactively with variable support.
Shareability: Easily share your prompt library via Git.

Key Features

✅ Organized Local Storage - Your prompts live in ~/.prompt-library (or project dir in dev). ✅ Git Integration - Optional syncing with your own remote repository. ✅ Multi-Model Support - Configure and switch between Anthropic & OpenAI models. ✅ Intelligent CLI - Interactive menus and powerful command-line options. ✅ Fragment System - Compose prompts from reusable text snippets. ✅ Variable Management - Define and manage global or prompt-specific variables. ✅ AI-Powered Metadata - Automatically analyze prompts to generate descriptions, tags, etc. (optional).

⚡ Quick Start

# 1. Fork the repository on GitHub first, then clone your fork
git clone https://github.com/YOUR_USERNAME/prompt-library.git

# 2. Install dependencies
cd prompt-library && npm install

# 3. Run in Development Mode (uses current directory for prompts)
npm run dev

# --- OR ---

# 3. Build and Install Globally (uses ~/.prompt-library)
npm run build
npm install -g .
prompt-library-cli menu # Start interactive menu

# 4. Initial Setup (IMPORTANT!)
# Run this after first launch to initialize your local library structure
prompt-library-cli setup

📋 Complete setup guide

See detailed instructions in Getting Started. You'll need to configure your AI provider API key using prompt-library-cli model.

🛠️ How It Works

The CLI primarily operates locally, managing prompts and fragments in your designated library directory (~/.prompt-library/repository by default).

Recommended Workflow (CLI-First)

Create Prompts: Use prompt-library-cli prompt create interactively. The CLI guides you through title, category, description, and content. AI analysis (optional, enabled by default) generates metadata (metadata.yml).
Create Fragments: Use prompt-library-cli fragment create to build reusable text snippets.
Manage & Execute: Use commands like prompt list, prompt search, execute, env, model, config to manage and run your library.
Sync (Optional): Use prompt-library-cli sync to push your local library changes to your configured remote Git repository.

Alternative Workflow (Git-Based)

Manually Create Files: Create prompt.md and metadata.yml files directly within category subdirectories inside ~/.prompt-library/repository/prompts.
Sync Database: Run prompt-library-cli sync (or the specific DB sync part if separated later) to update the local database with your filesystem changes.
Execute via CLI: Run prompts using prompt-library-cli execute.
Git Push: Manually commit and push changes using standard Git commands if you have a remote configured.

💡 Using GitHub Actions for Metadata (Optional)

If you maintain your prompt library in a dedicated GitHub repository (using the sync command or manually), you can enable GitHub Actions for automatic metadata updates:

Fork this repository or use your own containing the library structure.
Add your AI provider API key (ANTHROPIC_API_KEY or OPENAI_API_KEY) to your GitHub repository secrets.
When you push commits containing new or modified prompt.md files, the included GitHub Action (.github/workflows/update_views.yml) can:
- Analyze the prompt content using AI.
- Generate or update the corresponding metadata.yml file.
- Update the main README.md showcase section.
- Commit these changes back to your repository.

Note: The CLI workflow (prompt create, prompt update) handles metadata generation locally and does not require GitHub Actions. Actions are only needed if you prefer a Git-centric workflow for metadata.

🖥️ CLI Usage

The CLI provides comprehensive commands for managing your prompt library.

Main Commands

Usage: prompt-library-cli [options] [command]

Commands:
  menu                   (Default) Start the main interactive menu.
  execute [options]      Execute a specific prompt by ID or name, or via local files.
  prompt [subcommand]    Manage and browse prompts (list, search, create, update, delete, etc.).
  fragment [subcommand]  Manage reusable prompt fragments (list, create, update, delete, etc.).
  env [subcommand]       Manage global and prompt-specific environment variables (list, create, update, delete, read).
  model [options]        Configure AI model provider, model, and API key.
  config [options]       Manage CLI configuration settings.
  repository [options]   Manage the connection to your prompt library repository.
  sync [options]         Synchronize your local library with a remote Git repository.
  setup                  Initialize or configure the prompt library repository location.
  flush                  ⚠️ Reset all application data (prompts, fragments, history). Preserves config.
  generate-docs          Manually regenerate README files based on current prompts.
  help [command]         Display help for command.

Tip: Run prompt-library-cli <command> --help for detailed options on each command (e.g., prompt-library-cli prompt --help, prompt-library-cli prompt create --help).

Common Tasks

# Start interactive menu (recommended for exploration)
prompt-library-cli menu
# or just:
prompt-library-cli

# Initial setup
prompt-library-cli setup

# Configure AI Model (API Key, Provider, Model)
prompt-library-cli model

# Execute a prompt interactively (will prompt for variables)
prompt-library-cli execute -p <prompt_id_or_name>

# Execute non-interactively with variables
prompt-library-cli execute -p <id> --variable_one "value1" --another_var "value2"

# Execute with file content as input for a variable
prompt-library-cli execute -p <id> -fi input_data=./path/to/file.txt

# List all prompts
prompt-library-cli prompt list # or 'prompt ls'

# Search for prompts
prompt-library-cli prompt search "data analysis"

# Create a new prompt (interactive)
prompt-library-cli prompt create

# Update a prompt (interactive)
prompt-library-cli prompt update

# Delete a prompt (interactive, will ask for confirmation)
prompt-library-cli prompt delete

# List all fragments
prompt-library-cli fragment list

# Create a new fragment (interactive)
prompt-library-cli fragment create

# List environment variables
prompt-library-cli env list

# Set an environment variable (will create or update)
prompt-library-cli env update --key-value MY_API_URL=https://example.com/api # or 'env set ...'

# Set an environment variable to reference a fragment
prompt-library-cli env update --fragment COMMON_HEADER=templates/email_header

# Sync local changes with your remote Git repository (if configured)
prompt-library-cli sync

📚 Advanced CLI Usage

# --- Prompts ---
# List prompts sorted by ID
prompt-library-cli prompt list --id
# List only prompt categories
prompt-library-cli prompt list --categories
# View prompt details non-interactively
prompt-library-cli prompt read --prompt <id>
# Create prompt non-interactively from file
prompt-library-cli prompt create --title "My Agent" --category "coding" --description "Does things" --file ./prompt.md --directory my_agent --no-analyze
# Update prompt non-interactively with new content string
prompt-library-cli prompt update --prompt <id> --content "New prompt text..." --no-analyze
# Delete prompt non-interactively (force skip confirmation)
prompt-library-cli prompt delete --prompt <id> --force
# Refresh metadata for a specific prompt
prompt-library-cli prompt refresh-metadata --prompt <id>
# Refresh metadata for ALL prompts
prompt-library-cli prompt refresh-metadata --all
# List recent prompts (e.g., last 5)
prompt-library-cli prompt recent --limit 5
# List favorite prompts
prompt-library-cli prompt favorites

# --- Fragments ---
# List fragments non-interactively
prompt-library-cli fragment list # or 'fragment ls'
# Create fragment non-interactively
prompt-library-cli fragment create --category common --name header --content "# Header"
# Update fragment non-interactively from file
prompt-library-cli fragment update --category common --name header --file ./new_header.md
# Delete fragment non-interactively (force skip confirmation)
prompt-library-cli fragment delete --category common --name header --force

# --- Environment Variables ---
# Create/Update variable non-interactively (update is alias for set)
prompt-library-cli env update --key-value MY_VAR=abc
prompt-library-cli env update --fragment CONTEXT=common/project_details
# Delete custom variable non-interactively
prompt-library-cli env delete --name MY_VAR --force # Only deletes custom vars
# Read variable details non-interactively
prompt-library-cli env read --name MY_VAR
# Read variable value non-interactively
prompt-library-cli env read --name MY_VAR --value
# Read variable sources non-interactively (show prompt IDs)
prompt-library-cli env read --name MY_VAR --sources
# Read variable sources non-interactively (show prompt titles)
prompt-library-cli env read --name MY_VAR --sources --show-titles
# Get list output as JSON
prompt-library-cli env list --json

# --- Execution ---
# Inspect prompt variables without running
prompt-library-cli execute -p <id> --inspect # or -i
# Execute with JSON output (includes full response, timing, etc.)
prompt-library-cli execute -p <id> --json
# Execute with multiple file inputs
prompt-library-cli execute -p <id> -fi input1=./file1.txt -fi input2=./file2.log

# --- Configuration ---
# View specific config key non-interactively
prompt-library-cli config --key MODEL_PROVIDER
# Set config key non-interactively
prompt-library-cli config --key OPENAI_MODEL --value gpt-4o
# Reset config to defaults non-interactively
prompt-library-cli config --reset
# View config as JSON
prompt-library-cli config --json

# --- Repository & Sync ---
# View repository status details
prompt-library-cli repository # (Select 'View Status' or 'Change Branch' interactively)
# List pending local changes
prompt-library-cli sync --list
# Push pending local changes (requires commit message interactively or default non-interactively)
prompt-library-cli sync --push
# Push to a specific branch
prompt-library-cli sync --push --branch feature/my-prompts
# Reset (discard) all local changes non-interactively
prompt-library-cli sync --reset --force

📂 Prompt Showcase

Your AI Agent Collection

A starting point for building specialized AI agents.

Note: These are example prompts included with the library. Create your own collection tailored to your needs using prompt create!

🔹 Coding

Prompt	Description
Git Branch Name Generator	Generates optimized git branch names based on project context and user requirements
Git Commit Message Agent	Generates precise and informative git commit messages following Conventional Commits specification
GitHub Issue Creator	Creates comprehensive and actionable GitHub issues based on provided project information
Software Architect Code Reviewer	Generates comprehensive pull requests with architectural analysis and optimization suggestions
Software Architect Specification Creator	Creates comprehensive software specification documents based on user requirements
Software Architect Visionary	Analyzes user requirements and creates comprehensive software specification documents
Software Development Expert Agent	Provides expert, adaptive assistance across all aspects of the software development lifecycle.

🔹 Content Creation

Prompt	Description
Documentation Specialist Agent	Generates revolutionary software documentation using advanced AI techniques and industry best practices

🔹 Healthcare

Prompt	Description
Health Optimization Agent	Generates personalized, adaptive health optimization plans based on comprehensive user data analysis
Psychological Support and Therapy Agent	Provides AI-driven psychological support and therapy through digital platforms

🔹 Problem Solving

Prompt	Description
Problem Solving AI Agent	Generates expert networks and strategies to solve complex problems and achieve goals

🔹 Prompt Engineering

Prompt	Description
AI Assistant Architect	Conceptualizes innovative and feasible AI assistant designs for various domains
Prompt Engineering God	Crafts divine-tier prompts to maximize AI potential while adhering to ethical standards

🔹 Translation

Prompt	Description
Universal Translator Agent	Translates between any languages, modes of expression, or conceptual frameworks

💡 Pro Tip: Check out the Prompt Engineering category for prompts that help you create high-quality prompts

🚀 Getting Started

Step 1: Installation

# Fork the repository on GitHub first (optional, for backup/sharing)
# Clone the repository (or your fork)
git clone https://github.com/thibaultyou/prompt-library.git
cd prompt-library

# Install dependencies
npm install

# Build the project
npm run build

# Install the CLI globally
npm install -g .

# Verify installation
prompt-library-cli --version

Step 2: Initial Setup

Run the setup command. This creates the necessary directories (usually in ~/.prompt-library) where your prompts, fragments, database, and config will be stored locally.

prompt-library-cli setup

You'll be asked if you want to clone a remote repository, use an existing local folder, or create a default empty structure.

Step 3: Configure AI Provider & API Key

Configure the AI model you want to use. The CLI will guide you through selecting a provider (Anthropic/OpenAI) and entering your API key.

prompt-library-cli model

🔵/🟢 Where to get API Keys

Anthropic Claude: Get an API key from the Anthropic Console.
OpenAI GPT: Get an API key from the OpenAI Platform.

The key will be stored securely in the CLI's configuration file (~/.prompt-library/.config/config.json).

Step 4: Start Using!

You're ready to go!

# Start the interactive menu
prompt-library-cli

# Or use direct commands
prompt-library-cli prompt list
prompt-library-cli prompt create
prompt-library-cli execute -p <some_prompt_id>

🧩 Prompt Composition with Fragments

Fragments are reusable text snippets stored in the fragments/ directory, organized by category (e.g., fragments/common/disclaimer.md).

Using Fragments

Create: Use prompt-library-cli fragment create or manually create .md files.
Reference: Include fragments in your prompt.md files using {{VARIABLE_NAME}} syntax. Set the corresponding variable's value to Fragment: category/name using prompt-library-cli env update --fragment VARIABLE_NAME=category/name.
Manage: Use prompt-library-cli fragment commands (list, search, update, delete).

Examples: Check the included fragments/prompt_engineering/ directory for examples like behavior guidelines, formatting instructions, and safety rules.

⚙️ Metadata Generation

The CLI uses AI to help generate metadata (metadata.yml) when you create or update prompts using prompt create or prompt update.

Automatic Analysis: By default (--analyze flag is true), the CLI sends your prompt content to the configured AI model along with a system prompt (system_prompts/prompt_analysis_agent/prompt.md) to generate the title, description, tags, variables, etc.
Skip Analysis: Use the --no-analyze flag during create/update to skip the AI analysis and provide metadata manually or rely on basic extraction.
Refresh: Use prompt-library-cli prompt refresh-metadata (with --prompt <id> or --all) to re-analyze existing prompts.
Customization: You can modify the system prompt used for analysis by editing src/system_prompts/prompt_analysis_agent/prompt.md (requires rebuilding the CLI if installed globally).

🤝 Contributing

Contributions are welcome! Please refer to the project's contribution guidelines (if available) or submit issues and pull requests via GitHub.

📄 License

This project is licensed under the MIT License.

Name		Name	Last commit message	Last commit date
Latest commit History 115 Commits
.github/workflows		.github/workflows
fragments/prompt_engineering		fragments/prompt_engineering
prompts		prompts
src		src
test		test
.env.example		.env.example
.env.test		.env.test
.gitignore		.gitignore
.prettierrc		.prettierrc
.promptignore		.promptignore
LICENSE.md		LICENSE.md
README.md		README.md
build.js		build.js
eslint.config.mjs		eslint.config.mjs
jest.config.js		jest.config.js
jest.setup.ts		jest.setup.ts
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json
tsconfig.test.json		tsconfig.test.json

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Repository files navigation

🧠 Prompt Library CLI

📚 Table of Contents

🎯 Purpose & Features

Key Features

⚡ Quick Start

🛠️ How It Works

Recommended Workflow (CLI-First)

Alternative Workflow (Git-Based)

🖥️ CLI Usage

Main Commands

Common Tasks

📂 Prompt Showcase

Your AI Agent Collection

🚀 Getting Started

Step 1: Installation

Step 2: Initial Setup

Step 3: Configure AI Provider & API Key

Step 4: Start Using!

🧩 Prompt Composition with Fragments

Using Fragments

⚙️ Metadata Generation

🤝 Contributing

📄 License

About

Uh oh!

Uh oh!

Contributors 2

Uh oh!

Languages

License

thibaultyou/prompt-library

Folders and files

Latest commit

History

Repository files navigation

🧠 Prompt Library CLI

📚 Table of Contents

🎯 Purpose & Features

Key Features

⚡ Quick Start

🛠️ How It Works

Recommended Workflow (CLI-First)

Alternative Workflow (Git-Based)

🖥️ CLI Usage

Main Commands

Common Tasks

📂 Prompt Showcase

Your AI Agent Collection

🚀 Getting Started

Step 1: Installation

Step 2: Initial Setup

Step 3: Configure AI Provider & API Key

Step 4: Start Using!

🧩 Prompt Composition with Fragments

Using Fragments

⚙️ Metadata Generation

🤝 Contributing

📄 License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Uh oh!

Contributors 2

Uh oh!

Languages