Fluent Case Study

Why I Built Fluent

Useful advice often depends on more than the current conversation contains. Fluent brings the relevant information into the AI apps people already use.

The Product Thesis

Fluent works inside ChatGPT, Claude, Codex, OpenClaw, and other compatible AI apps. It brings in the information that matters for each question while the AI app handles the conversation and recommendation.

It is designed around several questions:

• Is this fact confirmed, inferred, or stale?
• Where did it come from?
• Should this AI app be allowed to use it?
• Can the assistant act on it?
• Did the change actually save?
• Can the user export or delete it?
• Can the same product work as both a hosted service and open-source software?

Current Product

• Meals can bring tastes and preferences, current grocery needs, and grocery spending into a meal-planning question.
• Style can bring closet items, photos, and a clothing budget into a purchase question.
• Budgets supplies optional grocery or clothing spending context when it is relevant to a Meals or Style decision.

Architecture

Fluent uses one MCP interface across ChatGPT, Claude, Codex, OpenClaw, and other compatible AI apps.

The main design choices are:

• typed shared and domain context
• source and certainty information for important details
• tenant-aware hosted storage
• approval before specific changes are saved
• verification that saved changes worked
• optional visual views that show the same saved information
• a shared contract across hosted and open-source runtimes

Fluent always works in text. Compatible apps can also show a visual Grocery List, Budget Setup, or Style Closet Manager.

Availability

Use Fluent online. There's no charge right now.

The open-source version is available for anyone who wants to inspect or run Fluent themselves.

What This Project Demonstrates

Fluent is an exercise in product architecture as much as implementation: deciding what an AI assistant should know, what it may change, what it must prove, and how those rules remain consistent across AI apps and setup options.