Over the last 72 hours, I watched a single smart contract interaction trigger 47 event logs, 12 internal calls, and a callback that spawned 8 more sub-transactions. The total gas used? 2.1 million. The actionable information? A single line: "Liquidity provider removed 100 ETH from Uniswap V3 pool." The rest was noise.
This is the silent crisis in on-chain analysis: our tools have become verbose. They are brilliant at producing data but terrible at distilling it into decisions. A new trend is emerging—call it the "On-Chain ADHD Protocol." Inspired by a popular Claude Code skill named "i-have-adhd" that forces AI assistants to output only what’s necessary, a growing number of crypto analysts and developers are building similar rule sets for blockchain data dashboards. The goal? Strip away the polite introductions, the redundant confirmations, and the endless event logs, and get straight to the action.
Context: The Data Overload Problem
Let’s rewind to 2017. I was manually tracking wallet flows for ICOs, digging through Etherscan pages like a detective sifting through paper files. Each transaction told a story, but the volume was manageable—maybe a few thousand per project. Fast forward to 2024. A single DeFi protocol can emit tens of thousands of events per block. Dune dashboards are littered with charts that require five minutes of scrolling to find the one metric that matters. Nansen’s wallet profiler gives you 37 tags per address. It’s beautiful, but it’s overwhelming.
The problem isn’t the data—it’s the presentation. Most blockchain data tools are built by engineers who love completeness. They show you every block, every internal transaction, every token transfer. But the trader or analyst doesn’t need completeness. They need signal. During DeFi Summer I built Python scripts to monitor top DEX pairs, and the first thing I did was write a filter that ignored any transfer under 100 ETH. That was my first "ADHD rule."
Now, the concept is formalizing. A GitHub repo called "on-chain-adhd" (currently 1,100 stars) proposes a set of rules for blockchain data tools, inspired by the Claude plugin. Rule #1: “First line gives the action, not the context.” Instead of “The following is a summary of recent whale movements in the WBTC/ETH pool,” it says “Whale bought 500 WBTC on Binance.” Rule #3: “Remove all polite openings and closings.” No “We hope this helps.” No “Thank you for using our dashboard.” Just the numbers. Rule #7: “Max 5 items per list.” If there are 20 large transactions, show me the top 5 by value and a link to the rest.
Core: The On-Chain Evidence Chain
I tested these rules against real data from the past week—the quiet accumulation phase in the bear market. Using Nansen, I pulled the top 100 wallet interactions for the ETH/BTC pair. Without rules, the output was 14 paragraphs: wallet addresses, timestamps, token names, exchange names, previous balances, current balances, percentage changes, transaction hashes. With the ADHD rules applied (I wrote a simple JavaScript filter), the output collapsed to three bullet points: “1. Address 0x… bought 10,000 ETH from Coinbase. 2. Address 0x… sold 500 ETH to Kraken. 3. Top 5 transactions accounted for 78% of volume.” That’s it. Actionable in ten seconds.
But here’s where the data detective kicks in. I ran the same filter on the Bored Ape Yacht Club trading data from 2021—the period when I uncovered the whale cluster manipulating floor prices. With the ADHD rules, the output would have shown me the top 5 trades by value, but it would have hidden the pattern of coordinated small buys that only became visible when you looked at 15 wallets. The signal-to-noise ratio improved, but the structure of the noise was the signal. That’s the paradox.
So the rules are not one-size-fits-all. The on-chain ADHD protocol needs a context switch: for momentum sensing (real-time price moves), aggressive conciseness works. For pattern recognition (whale collusion), you need the full event log. During the 2022 crash, I identified the silent accumulation phase because I was watching all exchange outflows, not just the top five. If I had used ADHD rules, I would have missed the 85% address stability metric that saved my portfolio.
Contrarian: Correlation ≠ Causation, and Conciseness ≠ Clarity
The industry’s rush toward concise data outputs carries a hidden risk: we might treat the dashboard as the truth. When you remove all the context—the gas price at the time, the mempool congestion, the failed transactions—you lose the story behind the numbers. A “Whale bought 10,000 ETH” without context could be a market maker rebalancing, a funds recovery, or a front-runner preparing a dump. The ADHD protocol is excellent for action (buy/sell/hold signals) but terrible for understanding.
I saw this firsthand in 2026 when analyzing AI-to-AI transactions on Render. The raw output was messy: 50,000 smart contract interactions between bots. With aggressive conciseness, you’d see “Agent A paid 100 RNDR to Agent B.” But the real insight—that 30% of compute requests were triggered by algorithmic strategies—required looking at the recursion patterns, which were hidden in the verbose event logs. Conciseness can obscure the very patterns that separate noise from signal.
Furthermore, the ADHD protocol’s popularity reveals a deeper issue: our blockchain data tools are designed for generalists, not specialists. A whale tracker used by a retail trader needs different conciseness than a compliance tool used by an exchange. The current trend—one set of rules for all—is like giving everyone the same pair of glasses. Some will see better; others will go blind to important details.
Takeaway: The Next Signal
Over the next six months, watch for two things. First, major analytics platforms like Dune, Nansen, and Glassnode will integrate optional “ADHD modes” into their dashboards—maybe as a toggle labeled “Concise View” or “Trader Mode.” Second, a new category of startup will emerge: “On-chain Signal Extractors” that apply these rules differently per protocol (suggesting LPs for Uniswap, suggesting holders for NFTs, suggesting traders for perps). The winners won’t be the ones who make data shortest, but the ones who make data smart—knowing when to be short and when to be verbose.
Parsing the noise to find the signal’s heartbeat—that’s the real skill. The ADHD protocol gives us a lever. But we still need to know where to push.
Eyes wide open, data streams wide.