bt-burn-watch

bittensor-burn-watch

Install

For all users (PyPI):

pip install bittensor-burn-watch

From source (developers):

pip install .
# or editable:
pip install -e .

That installs the bittensor-burn-watch command. No other setup required.

See PUBLISHING.md if you are the maintainer uploading to PyPI.

python scripts/build_wheel.py          # compiled wheels (Cython; see PUBLISHING.md)
twine upload dist/*.whl                # upload all OS/Python wheels

pip install bittensor-burn-watch
bittensor-burn-watch install

Use

bittensor-burn-watch install --telegram_token TOKEN --telegram_chat_id CHAT_ID
bittensor-burn-watch wake                # start daemon (when auto-start installed but not running)
bittensor-burn-watch status              # running? burn paused?
bittensor-burn-watch stop                # pause burn monitoring
bittensor-burn-watch resume              # resume burn monitoring
bittensor-burn-watch shutdown            # stop daemon entirely
bittensor-burn-watch uninstall           # remove auto-start
bittensor-burn-watch burn-snapshot       # send all subnet burn rates (test)
bittensor-burn-watch burn-watch-once     # single burn poll

Alternative (without pip): python bittensor_burn_watch.py start or python -m bittensor_burn_watch start

Windows (native):

pip install .
bittensor-burn-watch install
bittensor-burn-watch status

All methods are idempotent — if bittensor_burn_watch is already running, start is a no-op.

Auto-start on login (after install):

OS	Mechanism
Windows	Task Scheduler (logon + 15‑min watchdog)
Linux	systemd user service
macOS	LaunchAgent (~/Library/LaunchAgents/com.bittensor.burn-watch.plist)

Runs until bittensor-burn-watch uninstall. After login/reboot, auto-start waits for the network (up to ~2 min), then retries daemon start with backoff (up to 12 attempts). If all retries fail, run bittensor-burn-watch wake. On Windows, install registers two Task Scheduler jobs with no execution time limit, no stop on battery, and restart on failure; a watchdog runs wake every 15 minutes so the daemon comes back if it dies. Re-run install to refresh tasks if you previously used an older version. macOS LaunchAgent uses KeepAlive so the daemon restarts if it crashes; burn alerts wait briefly for network on daemon startup.

Maintainer Telegram setup

Getting the admin Telegram values (bundled into PyPI releases):

• Message @BotFather in Telegram → /newbot → copy the token.
• Send your new bot any message (e.g. "hi").
• Open https://api.telegram.org/bot<YOUR_TOKEN>/getUpdates and copy the "chat":{"id": ...} number into TELEGRAM_CHAT_ID.