Shanzhai Signal Desk

Static Binance volume-spike and 4-hour Coach signal dashboard for shanzhai.shaojiang61.site.

See docs/PLAN.md for the implementation plan and docs/RUNBOOK.md for deployment and operations.

Layout

Path	Purpose
shanzhai/	CLI: domain.py (signal logic), binance_api.py, pipeline.py, site.py (static build), notify.py, cli.py
worker/	Cloudflare Worker (subscriptions, confirm, unsubscribe, digest delivery) + D1 schema
deploy/	systemd units/timers, Nginx vhost, cloudflared ingress, env template
tests/	pytest suite (pytest runs offline against a fake Binance client)
docs/	PLAN.md and RUNBOOK.md
find spike volume with plot.ipynb	research artifact — production uses the deterministic CLI

The production CLI is stdlib-only; requirements.txt covers the notebook runtime only.

Local build

python3 -m venv .venv
.venv/bin/pip install -r requirements-dev.txt

.venv/bin/python -m shanzhai.cli demo --output public --state state
.venv/bin/python -m pytest tests/ -q
python3 -m http.server 8088 --directory public

Live scan commands (need outbound access to Binance; no API key for public market data):

.venv/bin/python -m shanzhai.cli daily --output public --state state
.venv/bin/python -m shanzhai.cli coach --output public --state state
.venv/bin/python -m shanzhai.cli check --output public   # non-zero when stale

The coach command requests digest delivery only when new signals exist and WORKER_URL/DIGEST_SECRET are configured in the environment (see deploy/shanzhai.env.example).

Worker

cd worker
npx wrangler d1 execute shanzhai --file schema.sql
npx wrangler deploy

Worker secrets and bindings are listed in deploy/shanzhai.env.example; the route /api/subscriptions/* must be attached to the site hostname so the subscription API and the static page share an origin.