Journal
Trade journal service for entries, trade logging, execution-quality analytics, and realised P&L tracking.
Part of FlintTrade — the open-source self-hosted trading software monorepo built with Python, React, TypeScript, and Rust.
Language: Python
Public surface
src/flinttrade_journal/trade_journal.py — journal-entry model and storage helperssrc/flinttrade_journal/order_analytics.py — execution-quality analyticssrc/flinttrade_journal/pnl_tracker.py — realised P&L calculations
(See the source for the full surface.)
Install
This package is part of the FlintTrade monorepo. Install via the workspace from the repo root:
uv pip install -e packages/services/journalIf you only want to use the package in isolation, the package's pyproject.toml,
Cargo.toml, or package.json lists its dependencies. The supported path is the
root workspace.
Tests
python -m pytest packages/services/journal/tests/ -v --import-mode=importlibFor the full test matrix, see the contributor guide at docs/DEVELOPER_GUIDE.md.
How this fits in
This package's role in the wider FlintTrade architecture is documented in docs/ARCHITECTURE.md. For end-user features it powers, see docs/USER_GUIDE.md.
Contributing
Contributions welcome. Please read contributing.md at the repo root before opening a pull request.
License
AGPL-3.0 — same as the parent repository. See LICENSE for the full text.
Engine
Five-layer order-admission safety, gated scheduled-strategy execution, scheduler, base strategy class, AST-guarded sandbox executor, bracket-order engine, mode guard, and reconcili
Screener
Option chain, OI analysis, PCR, max-pain, futures quadrant, portfolio Greeks, IV smile, fundamental screener, FII / DII tracker, and RRG calculator.