Research publishing system that turns public web and RSS evidence into traceable advisory reports.
English | 简体中文
Experimental. Use at your own risk. This project is not financial advice and does not guarantee any trading result. Its outputs are research evidence, not trading instructions.
QuantAdvisorResearch builds and publishes research-oriented advisory content from public web and RSS evidence. It is the deterministic composition layer of the advisory research system: it combines upstream evidence into advisory reports, market-confirmation and recommendation-review artifacts, and a static site, keeping each output traceable to its source. It is aimed at researchers and reviewers who need the published reasoning, not at live trading.
It does not place broker orders. It does not change allocations. It does not produce automatic trading instructions or auto-approvals, and its reports are not investment advice.
Requires Python 3.11 or newer.
python -m pip install -e '.[test]'
python -m pytest -qThe build and publish entry points print usage and required arguments when run
with --help:
qar-build-advisory-artifacts --help
qar-build-advisory-report --help
qar-publish-advisory-site --help| Variable | Required | Default | Purpose |
|---|---|---|---|
TELEGRAM_BOT_TOKEN |
No | none | Bot token used to send advisory notifications. |
TELEGRAM_CHAT_ID |
No | none | Chat that receives advisory notifications. |
NOTIFY_LANG |
No | zh |
Language used for notifications. |
Report and notification contracts live in docs/advisory_contract.md
and docs/notification_format.md.
src/ library code and command implementations
scripts/ command-line helpers
docs/ design notes and output contracts
examples/ sample inputs
data/ local working output
tests/ unit and contract tests
See CONTRIBUTING.md.
See SECURITY.md. Do not open a public issue for a vulnerability. Keep credentials, private data, and external service tokens out of Git and logs.
MIT. See LICENSE.