Scoring Methodology

How the Token Launch Safety Score (TLSS) works and what it measures.

The TLSS Framework

The Token Launch Safety Score is a 0–100 composite score derived from 22 weighted criteria across five research categories. It measures how safe, transparent, and structurally sound a token launch is—not whether it will go up in price.

Score Bands

Every scored project falls into one of four bands:

0–44
Critical
45–59
Caution
60–74
Fair
75–100
Strong

Categories Scored

Each report evaluates 22 criteria grouped into five categories:

Multi-Agent Consensus

Each report is produced by three independent research agents running in parallel:

1

Web & Docs Agent

Crawls public websites, documentation, whitepapers, and on-chain data.

2

Deep Document Agent

Performs detailed analysis of tokenomics papers, legal filings, and technical architecture.

3

Social Intel Agent

Evaluates community channels, team social presence, sentiment patterns, and engagement authenticity.

The final score is a consensus of all three agents. Where agents disagree, the report surfaces the disagreement explicitly rather than averaging it away.

Insufficient data = no score. If the pipeline cannot gather enough verifiable evidence to score a project honestly, no score is assigned. We never fabricate data or guess to fill gaps.

Calibration

TLSS scores are calibrated against a reference set of historical token launches with known outcomes. This ensures that a score of 75 today means the same thing it did six months ago, regardless of market conditions.

The TLSS is a research tool, not investment advice. A high score does not guarantee returns; a low score does not guarantee failure. Always do your own research.