Active accounts: 3 Strategies: AVAXUSDT(3x), BTCUSDT(3x), ETHUSDT(3x), LINKUSDT(3x), PAXGUSDT(3x), SOLUSDT(3x)
Account: mirat
| === DRY RUN MODE === | |
| ============================================================ | |
| JARVIS MULTI-ACCOUNT TRADER | |
| ============================================================ | |
| Active accounts: 3 | |
| Strategies: AVAXUSDT(3x), BTCUSDT(3x), ETHUSDT(3x), LINKUSDT(3x), PAXGUSDT(3x), SOLUSDT(3x) | |
| ------------------------------------------------------------ | |
| Account: mevlut | |
| ------------------------------------------------------------ | |
| Account: mirat |
Timestamp: 2026-03-04 09:38:18
Active accounts: mevlut, mirat, tugrul Strategies: AVAXUSDT(3x), BTCUSDT(3x), ETHUSDT(3x), LINKUSDT(3x), PAXGUSDT(3x), SOLUSDT(3x)
Fetched data for 6 symbols + 8 general endpoints
| name | description |
|---|---|
reddit-to-markdown |
Fetch a Reddit post and its comments as structured markdown.
Auto-triggers when the user's message contains a reddit.com URL pointing to
a submission (e.g. /r/sub/comments/id/...).
Example triggers: any message containing "reddit.com/r/*/comments/*".
|
Jarvis is a Django-based automated trading system for Binance Futures. It previously used a Genetic Algorithm (GA) to evolve trading strategies, but now uses an LLM (Claude Sonnet 4.6) as the sole decision-maker.
Key Facts:
Binance Global hesabına doğrudan TRY yatırılamaz. Önce Binance TR üzerinden para gönderilir:
User's thesis about the current BTC bear market (Feb 2026). This is the user's active trading thesis — reference in future sessions when discussing market conditions, SHORT positions, or training parameters.
The current BTC crash (98k → 60k) is a coordinated Wyckoff accumulation by whales/government, NOT a genuine bear market. The goal is to accumulate BTC cheaply before the US Strategic Bitcoin Reserve is filled.
| # Jarvis Emergency Mode - LLM Trading Prompt | |
| When emergency mode is active, this prompt is sent to an LLM subprocess | |
| with access to market data tools. The LLM fetches whatever data it needs | |
| and makes trading decisions autonomously. LLM-agnostic — works with any | |
| model that supports tool use or HTTP calls. | |
| --- | |
| ## Prompt (built per symbol) |