Redesigning TRAST: how we’re rebuilding the web3 trust layer from the ground up

4 min read4 days ago

*April 2025 marks a major milestone for TRAST: a complete database overhaul, a clean slate for entity data, and the official launch of TRAST v0.1-beta. This is the story of how we left legacy data behind and built something far more powerful in its place.*

A System That Grew Beyond Its Scope

When we first launched @TrastDotBot on Telegram, the goal was simple: help users verify and rate blockchain projects. But as the Web3 ecosystem evolved, so did user needs.

The community began using TRAST in unexpected and creative ways:

- Rating tokens, influencers and Telegram bots

- Reporting scams

- Tracking DAOs, exchanges, and launchpads

TRAST quickly became a go-to tool for adding new and emerging entities across the space — especially during early-stage project launches.

But our original database only had one entity type: projects.

Supporting these use cases meant duct-taping special cases and workarounds. Over time, the cracks started to show. The system was too rigid, too opinionated, and ultimately too fragile.

And as usage grew, we saw a recurring problem: duplicate and fragmented entries for the same real-world entity. A single influencer might exist under multiple entries — one for their $ticker, another for their @handle, another for their domain. There was no clear way to unify them.

Moving On from the Old Model

Instead of carrying over outdated structures and inconsistent data, we’ve turned the page. The previous system — while a valuable learning ground — wasn’t built for the scope TRAST is now aiming to cover.

This new foundation was created to be:

- Flexible enough to handle every Web3 entity type

- Scalable for future growth and integrations

- Auditable with transparent and verifiable claims

- Unified across all sources and identifiers

We’ve chosen long-term durability over short-term compatibility — and we believe users will feel the difference immediately.

The Core Idea: A Unified Web3 Entity Model

At the heart of the new system is a bold idea: every meaningful object in Web3 is a “profile”. Whether it’s a token, a DAO, a Telegram bot, or a crypto influencer — they’re all unified under a flexible, extensible model.

Every profile becomes the main entity that all relevant information links back to:

Reviews
Identifiers (contract address, domain, ticker, @tg-handle
Attributes (launch date, scam score, Twitter link)
Comments and community discussions
Audit history

And here’s the key design principle: whenever possible, all identifiers are resolved to a canonical main entity using their Telegram @handle.

Whether a project, user, DAO, or bot — if it has a Telegram presence, that becomes its anchor identity. If claimed by a user, the profile becomes linked to their account. If not yet claimed, the community can still create, tag, and verify it.

This approach:

Prevents fragmentation across identifiers
Encourages profile merging and ownership
Aligns with Telegram-native discovery and usage patterns

These changes are being rolled out gradually to ensure smooth adoption and backward compatibility.

How the new system works

We replaced the old schema with a modular, highly flexible one. Here are the key components:

1. Profiles table

One table to rule them all. Each profile has a username, entity_type (e.g. PROJECT, BOT, INFLUENCER), and shared fields like description, score, and verified.

2. Entity identifiers

Multiple identifiers (e.g. contract address, ticker, domain, @handle) can map to the same profile — all with their own verification status.

3. Flexible attributes

Rather than static columns, metadata is stored as key-value pairs. Examples:

launch_date: 2024-12-01
followers: 42000
scam_score: 87

4. Audit Trail

Every change is tracked: who changed what, and when. This transparency is vital for trust.

5. Claiming System

Users can claim profiles and become verified owners. This enables:

Direct profile management
Attribute and identifier verification
Built-in dispute resolution workflows

Clean Slate, Cleaner experience

The legacy data model has served its purpose — and we’ve left it in the past.

By starting from a clean slate, we’ve created a system that’s more structured, future-proof, and aligned with the evolving needs of the Web3 space.

From now on:

Every data point is auditable and verifiable
Every identifier is linked to a single profile
Every change is part of a transparent history

Introducing TRAST v0.1-beta

This relaunch also marks the beginning of formal versioning for TRAST.

v0.1-beta is rolling out now.

You can track upcoming updates via our release notes.

What’s new in this release:

Unified profiles for all entity types
Telegram-based identity resolution
Search powered by identifiers (not just usernames)
Flexible attribute metadata
Verified claiming & audit history

We’re inviting the community to explore, experiment, and help shape what comes next.

What you can do today

- Visit @TrastDotBot
- Search for any Web3 handle, domain, token or contract
- Create new profiles for anything that doesn’t yet exist
- Claim your profile and get verified
- Leave your first ratings and comments

This is just the beginning

TRAST is no longer just a project rating tool. It’s evolving into a decentralized, community-owned Web3 trust layer.

Future plans include:

- Public API for querying verified entity data
- Visual graphs of linked identities and scams
- ML-powered entity clustering and risk scores
- Profile reputation systems

We’re rebuilding TRAST not just for better code — but for a better Web3.

Join us.

— -

*Technical readers interested in the database architecture details can read our companion piece: [Under the Hood: The Technical Details of TRAST’s Database Redesign].*

— The TRAST Team