Onchain Analysis logo

Onchain Analysis

OrganizationPopular
HKUDS
onchain-analysis

On-chain data analysis — active addresses / whale tracking / TVL / DEX liquidity, interpretation and signal generation using on-chain valuation metrics such as MVRV / NVT / SOPR.

Overview

PublisherHKUDS
RepositoryVibe-Trading
Skill nameonchain-analysis
Stars
33.6K
Forks
5.5K
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by HKUDS on GitHub. Read the source before you install it.

Installation

Install the Onchain Analysis AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/HKUDS/Vibe-Trading.git /tmp/Vibe-Trading
mkdir -p .claude/skills
cp -r /tmp/Vibe-Trading/agent/src/skills/onchain-analysis .claude/skills/onchain-analysis
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Onchain Analysis in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Onchain Analysis on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Onchain Analysis is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

On-Chain Data Analysis

Overview

Use transparent public blockchain data for analysis, covering network activity, whale-behavior tracking, DeFi liquidity analysis, and on-chain valuation metrics. This provides crypto investing with data dimensions unavailable in traditional finance.

Network Activity Metrics

Core Activity Indicators

MetricMeaningBullish SignalBearish Signal
Active addressesUnique addresses interacting with the network each daySustained increase (network adoption rising)Sustained decline
New addressesAddresses appearing for the first time each dayAccelerating growth (new users entering)Shrinkage
Transaction countNumber of on-chain transactions per daySteady growthSharp decline
Transfer valueTotal value transferred on-chain per dayLarge-value transfer activityInactive
Activity / price ratioGrowth in active addresses vs growth in priceActivity growth outpaces price growthPrice rises but activity does not

Activity Analysis Framework

Healthy bull market:
  Price↑ + active addresses↑ + new addresses↑ = driven by real demand

Bubble signal:
  Price↑ + active addresses↓ or flat = capital-driven, not user-driven

Bottoming signal:
  Price↓ + active addresses bottom out and stabilize = speculators exit, real users remain

Historical Reference for BTC Active Addresses

PhaseActive Addresses (daily avg)BTC PriceMeaning
2020 bear-to-bull transition0.8-1.0 million$10k-$20kBottom stabilization
2021 bull market1.0-1.3 million$30k-$69kHealthy growth
2022 bear market0.8-0.9 million$16k-$30kPulled back but did not collapse
2024 bull market0.9-1.2 million$40k-$100kInstitution-driven

Whale Tracking

Whale Definition

TierBTC HoldingsETH HoldingsEstimated Count
Super whales>10,000 BTC>100,000 ETH~100
Large whales1,000-10,000 BTC10,000-100,000 ETH~2,000
Mid whales100-1,000 BTC1,000-10,000 ETH~15,000
Small whales10-100 BTC100-1,000 ETH~150,000

Whale Behavior Signals

Bullish signals:
1. Whales withdraw from exchanges -> intent to hold long term
2. Whale-wallet count rises -> institutions / large holders accumulating
3. Exchange BTC balance keeps falling -> lower available supply
4. Long-term holder (LTH) balances rise -> smart money buying

Bearish signals:
1. Large whale transfers into exchanges -> preparing to sell
2. Dormant addresses wake up and transfer -> early holders taking profits
3. Exchange balances surge -> sell pressure is about to be released
4. Miner balances fall and move to exchanges -> miner capitulation / profit taking

Large Transfer Monitoring

Thresholds to watch:
- BTC: single transfer > 500 BTC (about $50M)
- ETH: single transfer > 10,000 ETH (about $30M)
- USDT: single transfer > $50M

Transfer-direction analysis:
- Wallet -> exchange: potential selling (bearish)
- Exchange -> wallet: withdrawal to hold (bullish)
- Exchange -> exchange: arbitrage transfer (neutral)
- Wallet -> wallet: OTC trade (watch follow-up behavior)

DeFi Liquidity Analysis

TVL (Total Value Locked)

TVL = total value of assets locked in DeFi protocols

TVL analysis dimensions:
1. Total TVL trend: rising = capital flowing into the DeFi ecosystem
2. Cross-chain TVL: market-share changes among ETH vs Solana vs Arbitrum
3. Protocol TVL ranking: leading protocols such as Aave / Lido / Maker
4. TVL / market cap ratio: similar to asset-utilization rate in traditional finance

DEX Liquidity Metrics

MetricMeaningFocus
DEX volumeDaily decentralized-exchange volumeTrend in DEX / CEX ratio
Liquidity depthSize of AMM liquidity poolsBigger pools = lower slippage = better
LP yieldAnnualized return for liquidity providersAbnormally high = unsustainable
Impermanent lossOpportunity cost for LPsMore severe when volatility is higher

Stablecoin Liquidity

Stablecoin inflows = "dry powder" for the crypto market

Metrics to watch:
1. Changes in total USDT / USDC supply
2. Stablecoin balances on exchanges
3. Stablecoin mint / burn activity (Tether / Circle)
4. Stablecoin market-cap share (lower = higher risk appetite)

Signals:
- Heavy stablecoin minting -> capital preparing to enter
- Exchange stablecoin balances up -> buying power is accumulating
- Stablecoin share rising quickly -> risk-off market (capital rotating from coins into stablecoins)

On-Chain Valuation Metrics

MVRV (Market Value to Realized Value)

MVRV = market cap / realized cap

Realized cap = Σ(each UTXO × price at last movement)
             = sum of all holders' cost basis

| MVRV | Meaning | Historical Signal |
|------|------|---------|
| > 3.5 | Severely overvalued, large unrealized profits across holders | Historical top zone |
| 2.0-3.5 | Richly valued, most holders in profit | Mid-to-late bull market |
| 1.0-2.0 | Reasonable range | Normal market or early bull market |
| < 1.0 | Undervalued, most holders underwater | Bear-market bottom zone |

Signal logic: MVRV > 3.5 -> most holders have large unrealized gains -> strong incentive to sell -> potential top
              MVRV < 1.0 -> most holders are losing money -> unwilling to sell -> potential bottom

NVT (Network Value to Transactions)

NVT = market cap / daily on-chain transfer value

Similar to a PE ratio in traditional finance:
- High NVT: market cap is high relative to on-chain activity (overvalued or optimistic on future growth)
- Low NVT: market cap is low relative to on-chain activity (undervalued or value-like)

NVT Signal (improved):
NVT_Signal = market cap / MA(90, daily on-chain transfer value)
Use a 90-day moving average to smooth noise

| NVT Signal | Meaning |
|-----------|------|
| > 150 | Severely overvalued |
| 65-150 | Normal range |
| < 65 | Undervalued |

SOPR (Spent Output Profit Ratio)

SOPR = realized value of spent outputs / creation value of spent outputs

Simply put: on average, are the coins sold today being sold at a profit or a loss?

| SOPR | Meaning | Signal |
|------|------|------|
| > 1.05 | Sellers are realizing 5%+ profit on average | Profit-taking pressure |
| 1.0-1.05 | Small-profit selling | Normal |
| = 1.0 | Break-even | Key support / resistance |
| < 1.0 | Selling at a loss | Panic selling (bottom signal) |

In bull markets: a pullback to SOPR = 1.0 is a buy opportunity (cost-basis support)
In bear markets: a rebound to SOPR = 1.0 is a sell opportunity (cost-basis resistance)

Other On-Chain Valuation Metrics

MetricFormulaPurpose
Puell MultipleDaily miner revenue / MA(365, daily miner revenue)Miner-income cycle
Stock-to-Flowstock / annual productionBTC scarcity model
Reserve RiskHODL Bank / priceHolder confidence
Exchange balancetotal BTC held on exchangesSupply-side pressure

Analysis Framework

Composite On-Chain Score

Score each indicator from 1 to 5 and combine with weights:

| Dimension | Weight | Indicators |
|------|------|------|
| Valuation | 30% | MVRV, NVT |
| Activity | 25% | active addresses, new addresses |
| Capital flow | 25% | exchange balances, stablecoins |
| Whale behavior | 20% | whale holdings, large transfers |

Total score > 4.0: strongly bullish
Total score 3.0-4.0: bullish bias
Total score 2.0-3.0: neutral
Total score < 2.0: bearish / leaning bearish

Output Format

markdown
## On-Chain Analysis Report: BTC

### On-Chain Snapshot
| Metric | Current Value | Historical Percentile | Signal |
|------|--------|---------|------|
| MVRV | 2.1 | 65% | Elevated but not at a top |
| NVT Signal | 85 | 50% | Fair |
| SOPR(7d avg) | 1.02 | 55% | Slight profit-taking state |
| Active addresses | 950k/day | 45% | Relatively low |
| Exchange balance | 2.30M BTC | 30% | Low level (bullish) |

### Whale Activity
- Last 7 days: net withdrawal of +15,000 BTC (bullish)
- Large transfers: 3 transfers >1000 BTC into cold wallets
- Miners: holdings stable, no major outflows observed

### Composite Score: 3.5/5 (bullish bias)

### Conclusion
On-chain data is broadly bullish. MVRV is not yet in an extreme zone, exchange balances are low,
and whales continue to accumulate. But active addresses remain soft, so monitor whether this is an
"institution-driven bull market without retail participation". Maintain long exposure, but keep
position sizing controlled and avoid leverage.

Notes

  1. Data-source limitations: on-chain data is most reliable for BTC and ETH; data quality for other chains is uneven
  2. Entity identification is difficult: one entity can control multiple addresses, so whale analysis contains error
  3. UTXO vs account model: BTC (UTXO) analysis methods cannot be directly applied to ETH (account model)
  4. Missing Layer-2 data: many transactions occur on L2s (Arbitrum / Optimism), so L1 data is incomplete
  5. On-chain data lag: block confirmation takes time, so this is not suitable for short-term trading decisions
  6. Data acquisition: the built-in OKX data source provides candles / trade data, but on-chain data requires extra APIs (Glassnode / Nansen)
  7. Metric desensitization: as market structure changes (ETFization / institutionalization), historical thresholds may need adjustment

Frequently asked questions

What does the Onchain Analysis AI skill do?

On-chain data analysis — active addresses / whale tracking / TVL / DEX liquidity, interpretation and signal generation using on-chain valuation metrics such as MVRV / NVT / SOPR.

Why use Onchain Analysis on TypingMind?

Because you install it once and use it with any model. Onchain Analysis is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Onchain Analysis in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HKUDS/Vibe-Trading/tree/main/agent/src/skills/onchain-analysis. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Onchain Analysis?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Onchain Analysis?

As many as you like. As long as a model supports skills, you can use Onchain Analysis with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Onchain Analysis AI skill free?

Yes. It is published on GitHub by HKUDS under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇