Problem & Product

The Problem

People see friends and coworkers getting impressive value from AI tools, and they are often paying for those tools themselves, but they cannot figure out how to get the same results. They feel left behind -- a mix of FOMO, frustration, and mild inadequacy that grows as AI becomes more central to work and life.

Pain intensity: Moderate with emotional amplification and growing urgency. Not hair-on-fire -- nobody is losing money or getting fired over this. But the gap between "people who use AI well" and "people who don't" is widening, and the pressure to close it is building. Trovella lowers the activation energy, not the urgency. The urgency is building on its own; Trovella makes it easy enough to finally act on it.

How People Solve It Today

Current Approach	Why It Fails
Muddle through -- open ChatGPT, type vague questions, get mediocre answers	Assumes the user already knows what to ask. Reinforces the belief that "AI isn't that useful."
Watch YouTube / read articles -- passive consumption	Passive learning rarely changes behavior. High friction between watching and doing.
Ask a friend or coworker -- sporadic, unstructured	Actually the most effective channel, but not scalable and not available on demand.
Prompt libraries / template tools (FlowGPT, etc.)	Assumes the user already knows what they want to do. Building blocks without a blueprint.
Do nothing -- keep paying, use occasionally	The default. Comfortable inertia with a vague sense of leaving value on the table.

The common failure across all alternatives: They require the user to already know what they want to do. None put the user inside a guided experience where they are actually doing something fun or useful with AI.

The Product

Trovella is an AI capability layer that sits on top of users' existing LLM platforms and also provides its own experiences. It works in two directions:

1. Enhances Existing AI Tools (MCP Path)

A hosted MCP server that users connect to Claude, ChatGPT, Gemini, etc. It provides:

• Cross-platform memory -- context persists across sessions and across AI platforms
• Structured research capabilities -- multi-step research with plans, state tracking, and artifact persistence
• Curated skills (5-10 at launch) -- pre-built capabilities exposed through MCP that enhance what the user's AI app can do

Users' existing AI apps become smarter and more consistent without the user needing to do anything differently. See Research & Intelligence for the technical implementation.

2. Provides Its Own Experiences (App Path)

A suite of guided tools and creative mini-apps within Trovella itself:

Experience	What It Does	Why It Matters
AI image generation	Create pictures/paintings of friends and family in creative scenarios (e.g., family fighting dragons). Powered by Google Gemini.	Fastest "look what I just made" moment. High shareability. Proven demand. Primary hook.
Build-your-own adventure RPG	Single-player text-based RPG with AI storytelling and occasional generated images	Shares memory + image gen infrastructure. Real motivated users (Kyle's daughters). Fun engagement hook.
Preference engine	Tracks preferences, powers personalized recommendations (movies, restaurants, etc.)	Makes the app feel like it knows you. Foundation for future network effects.
AI tutoring	Interactive lessons, practice exercises, and skill progression for using AI effectively	Directly addresses the core problem. Turns passive frustration into active learning.
Professional document generation	AI-assisted PowerPoint, Word, and Excel from research results or prompts	High-value work output. Bridges "AI helped me think" to "AI helped me deliver."
Lightweight chat	Minimal in-app LLM chat	Necessary at launch but not the front door. Fallback for users who want direct chat.

These capabilities share common infrastructure (memory, image generation, preference tracking), making the breadth efficient to build. Each can launch at limited maturity and still be fun or useful.

Capability Discovery

Capability discovery is a primary design concern, not an afterthought. Users need to learn what the app can do without being overwhelmed. This IS the product for users who do not know what to do with AI -- without it, features will not be found.

The "Better Than" Test

Alternative	Why Trovella Wins
Doing nothing / muddling through	Trovella gives you things to do and guides you through them -- zero initiative required
YouTube tutorials / prompt guides	Passive learning vs. active doing. Trovella is the experience, not a lesson about the experience
Built-in suggestions in ChatGPT/Gemini	Those are generic and platform-locked. Trovella is personalized, cross-platform, and has curated experiences those platforms will never build
Prompt template tools (FlowGPT, etc.)	Templates are building blocks without a blueprint. Trovella is the finished room you walk into
Asking a knowledgeable friend	Trovella IS the knowledgeable friend, available 24/7, that remembers everything and keeps bringing you new things to try

The Core Architectural Insight

AI companies are currently subsidizing their consumer apps heavily with VC money -- $200/month in subscription fees can yield $2,500/month in API token equivalents. Trovella cannot afford to replicate this. The MCP-first architecture is an economic necessity: let users do heavy LLM work inside their subsidized apps, and Trovella enhances that work through memory, skills, and structure without bearing the token cost. Trovella's own chat and mini-apps handle lighter, targeted interactions where the cost is controlled.

Related Pages

• Target Users -- who specifically experiences this problem
• MVP Scope -- how the product definition maps to launch features
• Competitive Landscape -- where Trovella fits in the broader AI research tool market
• Research & Intelligence -- how the MCP research engine works
• Search & Retrieval -- how hybrid search powers the preference engine and research tools