CryptoOpsec🧭DYOR
Flight to fundamentals Β· open data Β· agent-ready

Qualify any crypto token on the dimensions that actually matter.

DYOR scores tokens on real revenue, durable tokenomics, and actual usage β€” asset-class-aware, so Bitcoin isn't judged like a DeFi app. Search any token, screen a universe, build a portfolio, or call it from your AI agent. Built on free, open data.

Everything DYOR does

πŸ”

Analyze any token

name Β· symbol Β· contract

Search by name, ticker, or contract address β€” it resolves the unified token across every chain. Get an asset-class-aware 0–1 score, A–D tier, gate flags, confidence + robustness, a market snapshot, every web/social link, a peer comparison, and a one-click analyst memo.

Open β†’
πŸ“Š

Tier screener

build Β· filter

A scored universe grouped into tier tabs (A→D). Build a fresh top-N by TVL in the background, or filter by asset class, minimum tier, real yield, and gate flags. Click any token to dive into its full analysis.

Open β†’
πŸ§ͺ

Portfolio Β· Barbell Β· Backtest

construction

Score a whole portfolio (tier + narrative exposure, flagged risks). Build the thesis' Barbell β€” a BTC anchor plus qualified, ungated satellites. Backtest whether the tiers actually predicted forward returns.

Open β†’
πŸ”₯

Narrative rotation

early signals

Which sectors are heating β€” AI, DePIN, RWA, gaming, privacy β€” ranked by momentum across 700+ CoinGecko categories. Spot capital rotation before price follows.

Open β†’
πŸ“–

Transparent methodology

not a black box

Every weight, tier threshold, hard disqualifier gate, asset-class profile, and metric definition β€” out in the open. Read exactly why a token scored the way it did.

Open β†’
🧭

Agent-callable (MCP)

for AI agents

DYOR ships a hosted MCP server β€” Claude, Cursor, Manus and other agents can call it as tools: analyze_token, screen_tokens, analyst_memo, score_portfolio, build_barbell, backtest. No install β€” point your agent at the URL and ask β€œis $TOKEN worth a look?” for an opinionated, gated read.

claude mcp add --transport http dyor https://dyor.cryptoopsec.com/mcp
API & MCP docs β€” endpoints, connection, tools β†’

How it works

Step 1
Ingest
Live data from free/open sources β€” DefiLlama, CoinGecko, CryptoRank, Ethplorer, Santiment, GitHub, Sourcify.
Step 2
Classify
Type each token β€” DeFi, L1, monetary, memecoin, stablecoin β€” and judge it on what matters for it.
Step 3
Measure
Derived metrics: P/F, P/S, FDV/MCAP, token-sink, unlock overhang, holder concentration, growth.
Step 4
Score
Normalize across same-class peers, weight by domain, then gate β€” hard red flags cap or zero the score.
Step 5
Rank
Map the 0–1 score to a tier β€” A (high conviction) β†’ D (avoid) β€” with a confidence + robustness read.

Judged on their own terms

Each token is classified and scored with a class-appropriate profile β€” β€œno protocol revenue” is fatal for a DeFi app but a non-issue for Bitcoin.

DeFi protocol
Fees, revenue, TVL, value accrual.
L1 / platform
Ecosystem TVL, adoption, dev activity.
Monetary
Scarcity + accumulation β€” no revenue expected.
Memecoin
Distribution, liquidity, social attention.
Stablecoin
Adoption + distribution; not a price play.
🚫 The gate

Hard disqualifiers cap or zero a score so a flaw can't be averaged away β€” unverified contract, extreme FDV/MCAP, or a dead token (no commits 6mo+, ~99% off ATH, near-zero volume).

🎯 Confidence + robustness

Every score says how complete the data is and whether the tier survives re-weighting β€” so a thin or fragile call is labelled, not hidden.

🟒 Open & free

No paywall, no black box. Built entirely on free/open data, with a transparent, inspectable methodology and an open API.

Search any token β€” by name, symbol, or contract address.
Research aid, not financial advice.
Start analyzing β†’