Crypto / agent fleet

An army of agents that never stop reading crypto Twitter

Scan X around the clock. Build a thesis on a coin from the very first time anyone writes about it, and keep updating it until the story dies. Work out where momentum actually is, and eventually place the trade without a human in the loop.

Phase 1 updated no live capital

The idea, as written down

Agents that help with: - Scanning Twitter all the time. Writing theses about coins from infancy, so from the very first time anything is written about them, all the way to the end. And then predicting where the momentum is and what should be bought. - Basically an army of agents that are chronically online, have perfect memory, and can analyse and monitor a lot of things at the same time. - Then let them trade automatically.

Everything below is that note turned into something buildable, with the parts that are hard stated as hard rather than hidden.

The correction that shaped the design. A memecoin today is a ponzi with a chart. There is no cash flow and no product, so the only thing that pays you is somebody arriving after you did. That makes this a timing problem in both directions: you have to be in front of the crowd on the way in and in front of it on the way out. Every agent below is built around that, and it is why the scorer grades accounts on lead time rather than on returns. An account that posts winners but posts them late is not a weak buy signal - it is the crowd arriving, which is the sell signal.

Where it stands

The fleet - six roles

Not one agent doing everything. Six narrow jobs, each cheap to run and each independently checkable, because a wrong answer in one of them has to be visible instead of blended into a confident summary.

1 - Scout

Reads the timeline

Pulls posts from a watchlist of accounts in batches, plus keyword and cashtag sweeps for tickers nobody on the list has mentioned yet. Extracts ticker, contract address, chain and whether the post is a call, a mention or paid promo.

2 - Historian

Writes the thesis from infancy

The moment a token is seen for the first time, it gets a file. Every later mention is appended with who said it, when, and how big their audience is. The file is the coin's biography: first sighting, who picked it up, when the tone changed, when it died. Nothing is ever overwritten.

3 - Scorer

Grades the accounts

Backtests every account on its own past calls. Median return, hit rate, and the only number that really matters: how early they are relative to everyone else on the same coin. An account that is loud but late is worse than useless, it is a sell signal.

4 - Analyst

Checks the chain

Before you ever see an alert: pool age, liquidity, is the LP burned, what the dev wallet holds, top ten holder concentration, and whether the contract can be minted or frozen. Kills most candidates outright. This is the filter that stops the fleet becoming a rugpull delivery service.

5 - Signal

Watches the spread rate, both ways

The CYBERLEEK pattern: not one caller, but more and more accounts talking about the same thing while the gaps between mentions get shorter. Fires when two or more scored accounts converge on one contract inside a window and the rate is accelerating - not on any single tweet.

And it fires again on the way out. The same counter that says get in says get out: the moment the known-late accounts and the automated whale feeds start posting the ticker, the crowd has arrived and there is nobody left to sell to. On CYBERLEEK that was whalewatchalert and dexvralisting, half an hour behind the wave. Exit alert, not a buy alert.

6 - Executor

Places the trade

Last, and switched off. Fixed size per call, a hard daily cap, an automatic exit ladder, and a separate wallet holding nothing but the trading budget. Cannot run until the five agents above have a track record on paper.

Build order

  1. Score the accounts running now

    Build the watchlist and backtest it. No list exists yet, so the fleet seeds its own: work backwards from CYBERLEEK, CATE and EGG and find who mentioned each of them first, then second, then third. Those accounts become candidates, and their own history gets scored the same way. Deliberately slow - batched scrapes, spread over days, so nothing gets hammered. Output: a ranked list of who is genuinely early.

  2. Theses from infancy needs a scored list

    Every token the scouts see gets a file from first sighting. Runs read-only for a few weeks so we can look back and ask: at what point in this file was it obvious? That question cannot be answered on historical data alone, because dead tokens lose their pool history and quietly drop out of any backtest, flattering the result.

  3. Live alerts needs signal in phase 2

    Convergence plus safety check, pushed to WhatsApp within minutes with liquidity, market cap, pool age and the thesis so far. Every alert is logged with entry price at alert time, so the scoreboard on this page is not curated afterwards.

  4. Paper trading gate: alerts must beat the benchmark

    Fixed size into every alert, exits by rule, no discretion. Compared against the benchmark below. If paper does not clear it over a meaningful number of trades, the project stops here, and that result gets published on this page too.

  5. Automatic execution gate: paper must be profitable after fees

    Separate wallet, funded with the trading budget only, seed used nowhere else. Hard per-trade and per-day limits. Never the main wallet - whatever key sits on that machine is a key the machine can spend.

The benchmark to beat

One caller was backtested properly before any of this was built. Everything the fleet produces gets measured against it - if a whole agent fleet cannot beat one guy's timeline, it is not worth running.

StrategyReturn

Full call-by-call breakdown: stitch-calls-concept.jouwidealewebsite.nl

Results - account scores

Results - coin dossiers

Results - alerts fired

Results - paper trades

What can go wrong

The base rate is brutal. On the one caller measured so far, the median call is worth half your money today. Reading more accounts does not fix that. It only produces more of them.

Cost and load

Scraping spend

Read live from the billing API, not from a tally the system keeps about itself. The loop checks it before every single scrape and refuses to start one that would cross the line - at the ceiling it stops scraping and keeps publishing, it does not slow down and carry on.

Scraping

Roughly 0.0002 dollar per post, and only ever batched - one request covers four accounts at once. Polling accounts separately is billed per query and would run into hundreds a day, so the fleet never does it.

Pace

One batch every six hours, one at a time, with a wait between every price lookup. Roughly sixteen accounts a day. Nothing here is time-critical, and the same machine also runs the CRM, WhatsApp and the NFT watchers - those come first.

Load on the box

A single node process under PM2, asleep almost all the time. Price history comes from a free endpoint at one request every 2.5 seconds, well under its limit. Stop it any time with pm2 stop crypto-agents.

Log

DateChange