Product walkthrough
What we checked We read the product source at dmitriy718/tradeflows-pro@2e5540a (2025-12-16), in an audit dated 2026-09-17. That's the source code only — we couldn't confirm what's deployed at app.tradeflows.net (it answered 522, origin unreachable, at the last check on 2026-09-17).
We can't link a live app, because we can't verify what's deployed. What we can show honestly is the workflow itself, worked on synthetic data that never changes — so what you see is exactly what was confirmed, no more. The individual capabilities behind it are listed on verified capabilities; the environment the app needs is on requirements.
The walkthrough: the verified workflow, worked in the open
This walkthrough steps through one thing the TradeFlows app does — load market data, draw a chart, compute indicators — using a fixed synthetic series for the fictional symbol EXAMPLE. Walkthrough version tfnet-walkthrough-v1. Nothing is uploaded, requested or
stored; the data never changes, so your result is identical to everyone
else's. It demonstrates what the software does — it does not recommend
buying, selling or holding anything.
Step 1 of 5: The sample input
Step 1: The sample input
Everything in this walkthrough runs on the fixed series below: 16 invented daily candles for a fictional symbol, EXAMPLE. It is synthetic practice data — not market data, and it never changes, so your result and everyone else’s are identical. You can download it as the sample input at the end.
release evidence: Synthetic scenario — no source equivalent; labeled practice data| Date | Open | High | Low | Close |
|---|---|---|---|---|
| 2026-09-01 | 98.20 | 99.00 | 97.90 | 98.50 |
| 2026-09-02 | 98.60 | 99.60 | 98.30 | 99.20 |
| 2026-09-03 | 99.00 | 99.30 | 97.50 | 97.80 |
| 2026-09-04 | 97.90 | 99.10 | 97.70 | 98.90 |
| 2026-09-07 | 99.20 | 100.40 | 99.00 | 100.10 |
| 2026-09-08 | 100.30 | 101.80 | 100.00 | 101.40 |
| 2026-09-09 | 101.20 | 101.70 | 100.30 | 100.60 |
| 2026-09-10 | 100.90 | 102.60 | 100.80 | 102.30 |
| 2026-09-11 | 102.50 | 103.40 | 102.10 | 103.10 |
| 2026-09-14 | 102.80 | 103.00 | 101.90 | 102.20 |
| 2026-09-15 | 102.50 | 104.10 | 102.40 | 103.80 |
| 2026-09-16 | 104.00 | 105.40 | 103.90 | 105.10 |
| 2026-09-17 | 104.90 | 105.30 | 104.20 | 104.60 |
| 2026-09-18 | 104.90 | 106.50 | 104.80 | 106.20 |
| 2026-09-21 | 106.40 | 107.30 | 106.10 | 107.00 |
| 2026-09-22 | 107.20 | 108.70 | 107.00 | 108.40 |
Step 2: How the app gets data
In the app's source code, it reads public market data through a separate read-only proxy service; provider keys live on that service, never in your browser, and the source has data routes only — no order routes. This page performs no request; the frame shows the documented pattern as an illustration. We don’t know which version of the proxy is deployed, which is one reason live charts may not load in the app today.
release evidence: src/services/MarketDataAggregator.ts; tf-proxy worker source — read-only data routes onlyGET /marketdata/EXAMPLE/daily (illustration — no request is made) → read-only proxy service → public market data 200 OK — 16 daily candles source: synthetic series embedded in this page (tfnet-walkthrough-v1)
Step 3: The series as a chart
Each candle summarizes one session: the body spans open to close; the wicks reach the high and low. Amber candles closed at or above their open; gray candles closed below. The same numbers are in the step-1 table — the chart is only a shape drawn over them.
release evidence: lightweight-charts dependency; chart components — checked in sourceStep 4: Two indicators, computed in the open
The app computes indicators in Web Workers — background scripts inside your browser. Here is the simplest family, worked in the open: a fast 4-session and a slow 8-session simple moving average of the closing prices — add the last N closes, divide by N. Those two numbers are the whole "indicator": arithmetic on history.
release evidence: src/ Web Worker signal modules — indicator arithmetic in browser background scriptsfast average — 4-session simple moving average of closes (last 4 sessions: 2026-09-17, 2026-09-18, 2026-09-21, 2026-09-22) (104.60 + 106.20 + 107.00 + 108.40) / 4 = 106.55 slow average — 8-session simple moving average of closes (last 8 sessions: 2026-09-11, 2026-09-14, 2026-09-15, 2026-09-16, 2026-09-17, 2026-09-18, 2026-09-21, 2026-09-22) (103.10 + 102.20 + 103.80 + 105.10 + 104.60 + 106.20 + 107.00 + 108.40) / 8 = 105.05 Latest close 108.40 is above the fast average. The fast average is above the slow average. That is the entire computation. It describes the past of this synthetic series; it recommends nothing.
Step 5: Outcome and exports
You have seen input → data path → chart → indicators, with the working shown. Take the sample input (CSV) and the expected output (text) with you and check the arithmetic anywhere you like. Boundary, restated: this walkthrough demonstrates what the software does. It does not recommend buying, selling or holding anything, and no result here describes a real instrument.
release evidence: Walkthrough output — deterministic; identical for every visitorResult: fast average 106.55 (4 sessions), slow average 105.05 (8 sessions); the fast average is above the slow one — computed in your browser, shown in full above.
The CSV is the same synthetic series with a provenance comment header; the text file is the result any correct re-computation must reproduce — checking the arithmetic in a spreadsheet is the point of the exercise.
Text version (the same demonstration, start to finish)
Everything the interactive steps show, as one document. This is also the static equivalent for anyone who prefers not to use the stepper.
TradeFlows product walkthrough (tfnet-walkthrough-v1) Synthetic data: symbol EXAMPLE is fictional; the series below is invented practice data, not market prices. Step 1 of 5: The sample input Evidence: Synthetic scenario — no source equivalent; labeled practice data Everything in this walkthrough runs on the fixed series below: 16 invented daily candles for a fictional symbol, EXAMPLE. It is synthetic practice data — not market data, and it never changes, so your result and everyone else’s are identical. You can download it as the sample input at the end. Step 2 of 5: How the app gets data Evidence: src/services/MarketDataAggregator.ts; tf-proxy worker source — read-only data routes only In the app's source code, it reads public market data through a separate read-only proxy service; provider keys live on that service, never in your browser, and the source has data routes only — no order routes. This page performs no request; the frame shows the documented pattern as an illustration. We don’t know which version of the proxy is deployed, which is one reason live charts may not load in the app today. Step 3 of 5: The series as a chart Evidence: lightweight-charts dependency; chart components — checked in source Each candle summarizes one session: the body spans open to close; the wicks reach the high and low. Amber candles closed at or above their open; gray candles closed below. The same numbers are in the step-1 table — the chart is only a shape drawn over them. Step 4 of 5: Two indicators, computed in the open Evidence: src/ Web Worker signal modules — indicator arithmetic in browser background scripts The app computes indicators in Web Workers — background scripts inside your browser. Here is the simplest family, worked in the open: a fast 4-session and a slow 8-session simple moving average of the closing prices — add the last N closes, divide by N. Those two numbers are the whole "indicator": arithmetic on history. Step 5 of 5: Outcome and exports Evidence: Walkthrough output — deterministic; identical for every visitor You have seen input → data path → chart → indicators, with the working shown. Take the sample input (CSV) and the expected output (text) with you and check the arithmetic anywhere you like. Boundary, restated: this walkthrough demonstrates what the software does. It does not recommend buying, selling or holding anything, and no result here describes a real instrument. Data table (date, open, high, low, close): 2026-09-01 98.20 99.00 97.90 98.50 2026-09-02 98.60 99.60 98.30 99.20 2026-09-03 99.00 99.30 97.50 97.80 2026-09-04 97.90 99.10 97.70 98.90 2026-09-07 99.20 100.40 99.00 100.10 2026-09-08 100.30 101.80 100.00 101.40 2026-09-09 101.20 101.70 100.30 100.60 2026-09-10 100.90 102.60 100.80 102.30 2026-09-11 102.50 103.40 102.10 103.10 2026-09-14 102.80 103.00 101.90 102.20 2026-09-15 102.50 104.10 102.40 103.80 2026-09-16 104.00 105.40 103.90 105.10 2026-09-17 104.90 105.30 104.20 104.60 2026-09-18 104.90 106.50 104.80 106.20 2026-09-21 106.40 107.30 106.10 107.00 2026-09-22 107.20 108.70 107.00 108.40 TradeFlows walkthrough — expected output (tfnet-walkthrough-v1) Input: synthetic series for fictional symbol EXAMPLE (16 sessions) — not market prices. Fast average (4 sessions, 2026-09-17, 2026-09-18, 2026-09-21, 2026-09-22): (104.60 + 106.20 + 107.00 + 108.40) / 4 = 106.55 Slow average (8 sessions, 2026-09-11, 2026-09-14, 2026-09-15, 2026-09-16, 2026-09-17, 2026-09-18, 2026-09-21, 2026-09-22): (103.10 + 102.20 + 103.80 + 105.10 + 104.60 + 106.20 + 107.00 + 108.40) / 8 = 105.05 Latest close 108.40 is above the fast average. The fast average is above the slow average. That is the entire computation. It describes the past of this synthetic series; it recommends nothing.