FleaBot is a data-driven toolkit for Escape from Tarkov players. It started as a personal side project to answer one question: "Am I actually making money on the flea market?"
Tarkov's economy is one of the most complex in any game. Prices fluctuate with patches, wipes, trader resets, and player behaviour. The in-game flea market gives you a snapshot, but no history, no trends, and no way to know if the 15 000₽ item you listed this morning is going to be worth 8 000₽ by tonight.
We built FleaBot to fix that. It aggregates price history from the open tarkov.dev community API, parses local game logs to extract raid and queue data, and surfaces all of it in a dashboard that tells you exactly how your sessions are going.
The free tier exists because we believe basic market data should be free. The Pro tier funds the servers, development time, and the price-alert infrastructure that fires within seconds of a price change.
FleaBot was built out of frustration with not having data-driven tools for the flea market. Every feature exists because someone on the team wanted it in their own session.
Feature requests come from Discord. Accuracy metrics are public. When our predictions are wrong we say so — the /market/accuracy page shows live hit rates, not cherry-picked numbers.
Our price and trend data is freely available via API with no auth required. If you want to build your own tool on top of our data, the embed docs are at /embed.
Where the numbers come from and how we process them
All item prices are pulled from the tarkov.dev community GraphQL API — an open, community-maintained database of every item in Escape from Tarkov. We do not scrape the game client.
Market prices refresh every 5 minutes via the tarkov.dev API. Queue time data is aggregated in real time as users submit game logs, giving crowd-sourced accuracy.
The desktop client reads EFT's local game logs from your machine. Logs are parsed client-side first; only aggregated stats (raid count, queue times, death causes) are sent to our servers — never raw log text.
Trend forecasts are based on rolling 7-day and 30-day price history for each item. Trader reset cycles and wipe-day patterns are factored in as known seasonal signals.
Queue wait times are averaged across opted-in users per map, per hour. Individual times are never stored; only the aggregate is exposed. Data older than 24 hours is excluded from the live display.
We store the minimum data needed to provide the service. No gameplay footage, no keystrokes, no inventory screenshots. Full detail in our Privacy Policy.
Read moreWe use cookies to enhance your experience. Essential cookies are required for the site to function. Optional cookies help us improve our services. Read our Privacy Policy