This paper presents a comprehensive longitudinal study of the emerging Large Language Model (LLM) app ecosystem, focusing on two GPT app stores: GPTStore.AI and the official OpenAI GPT Store. The study aims to understand the evolution, landscape, and vulnerabilities of these apps over a five-month period. Key findings include: 1. **User Enthusiasm**: User enthusiasm for GPT apps consistently rises, while creator interest plateaus within three months of GPTs' launch. 2. **App Usage Distribution**: The app usage follows a power law distribution, with a small number of apps accounting for most user interactions. 3. **Configuration Vulnerability**: Nearly 90% of system prompts can be easily accessed due to widespread failure to secure GPT app configurations, leading to plagiarism and duplication. 4. **Plagiarism Issues**: Notable instances of plagiarism within the GPT app store have raised copyright issues, with over 705 apps sharing nearly identical names and descriptions. The study recommends that app stores implement continuous monitoring to notify creators of potential leaks and duplication issues, introduce financial incentives to attract creators, and enhance protective mechanisms to secure app configurations. Creators are advised to protect their configurations and consider moving sensitive data to external APIs. Users are encouraged to provide feedback and ratings to improve the quality and reliability of apps in the ecosystem.This paper presents a comprehensive longitudinal study of the emerging Large Language Model (LLM) app ecosystem, focusing on two GPT app stores: GPTStore.AI and the official OpenAI GPT Store. The study aims to understand the evolution, landscape, and vulnerabilities of these apps over a five-month period. Key findings include: 1. **User Enthusiasm**: User enthusiasm for GPT apps consistently rises, while creator interest plateaus within three months of GPTs' launch. 2. **App Usage Distribution**: The app usage follows a power law distribution, with a small number of apps accounting for most user interactions. 3. **Configuration Vulnerability**: Nearly 90% of system prompts can be easily accessed due to widespread failure to secure GPT app configurations, leading to plagiarism and duplication. 4. **Plagiarism Issues**: Notable instances of plagiarism within the GPT app store have raised copyright issues, with over 705 apps sharing nearly identical names and descriptions. The study recommends that app stores implement continuous monitoring to notify creators of potential leaks and duplication issues, introduce financial incentives to attract creators, and enhance protective mechanisms to secure app configurations. Creators are advised to protect their configurations and consider moving sensitive data to external APIs. Users are encouraged to provide feedback and ratings to improve the quality and reliability of apps in the ecosystem.