finbot

Description

Features

• Discord, Slack and Telegram support
• Sharesight API features:
  • Trade notifications
  • Portfolio performance
  • Holdings list
• Yahoo Finance data for Sharesight holdings:
  • Analyst consensus changes
  • Earnings date reminders
  • Ex-dividend date warnings
  • Highly shorted stock warnings (AU, US)
  • Interday and premarket price movements over a defined threshold
• Interactive chat commands (Slack & Telegram):
  • All the above reports
  • Price history graph
  • Stock lookup with price/valuation related stats and warnings
  • Company profiles
  • Shared watch list
  • List stocks (holdings + watchlist) by P/E, Forward P/E, PEG, market cap, or Yahoo analyst consensus

Requirements

• Most features require a Sharesight paid plan, preferably with automatic trade imports, and an API key
• Discord/Slack webhooks / Telegram bot user
• Interactive commands require a public web server with domain name and matching certificate to proxy chat commands to the bot.
• Python ≥ 3.10
• Python modules:

datetime python-dotenv requests gevent pandas matplotlib numpy

Supported commands

Where:

• percent is a number followed by '%'
• period is a number followed by 'd' 'w' 'm' or 'y'
• portfolio is a Sharesight portfolio name
• symbol is a Yahoo Finance symbol

.symbol
.dividend [period|symbol]
.earnings [period|symbol]
.history symbol
.holdings portfolio
.marketcap [symbol|top|bottom]
.performance [period] [portfolio]
.premarket [percent|symbol]
.price [percent|symbol] [period]
.profile symbol
.rating [strong buy|buy|hold|underperform|sell]
.session [percent|symbol]
.shorts [percent|symbol]
.trades [period] [portfolio]
.value [pe|forward pe|peg|bottom pe|bottom forward pe|bottom peg|negative forward pe|negative peg]
.watchlist [add|del symbol]

Alternative syntax:

@botname <symbol>
@botname dividend [period|symbol]
@botname earnings [period|symbol]
@botname history symbol
@botname holdings [portfolio]
@botname marketcap [symbol|top|bottom]
@botname performance [period] [portfolio]
@botname premarket [percent|symbol]
@botname price [percent|symbol] [period]
@botname profile symbol
@botname rating [buy|hold|underperform|sell]
@botname session [percent|symbol]
@botname shorts [percent|symbol]
@botname trades [period] [portfolio]
@botname value [pe|forward pe|peg]
@botname watchlist [add|del symbol]

Trade notifications are peformed by polling the Sharesight trades API from a cron job, and notifying your configured chat networks of any new trades. Thus, it works best if your trades are auto-imported into Sharesight through its broker integrations, and if your environment has persistent storage so that the bot can keep track of known trade IDs between runs.

The other various reports can also run from cron (e.g. daily or weekly), or on demand through the interactive bot. They query the Yahoo Finance API for stock data based on current holdings across your Sharesight portfolios, your friends' Sharesight portfolios, plus a custom watch list. Depending on how they're triggered, they will either report to all configured chat networks, or reply to the chat which triggered them.

The interactive bot component requires you to host a web service on a domain with a trusted certificate. It subscribes to push updates from native Slack apps / Telegram bots, and reacts to certain regex seen in chat. It provides:

• Stock lookup (financials, company profile, price charts)
• Group maintainable watch list, picked up by the various reports
• Listing of portfolios and portfolio performance
• Listing of holdings sorted by various attributes
• Running the other stock reports on demand

Reports

Stock lookup

The stock lookup returns various stats relevant to a stock's valuation, growth and risk factors.

Emoji legend

• The colored circle next to the symbol at the top indicates the current market state:
  • 🔴 market closed
  • 🟠 pre/post-market
  • 🟢 normal trading
• The warning symbol ⚠️ indicates possible risk factors. This is not intended to replace due diligence, but to non-exhaustively highlight where additional diligence is due. Some risk factors include:
  • Negative earnings ⚠️
  • Earnings in decline ⚠️
  • Negative cashflow ⚠️
  • High short interest ⚠️
  • High forward P/E ratio ⚠️
  • Net insider selling ⚠️
  • High debt-to-equity ratio ⚠️
• The arrow grid indicates the directional change in the company earnings and revenue for each reporting period. A red arrow indicates the absolute earnings was negative:
  • 🔼 earnings/revenue increased from the previous period
  • 🔽 earnings/revenue decreased from the previous period
  • 🔺 earnings increased from the previous period but remained negative
  • 🔻 earnings decreased from the previous period and was negative
  • ▪️ earnings/revenue matched the previous period
  • ❌ data is missing for this period
• Using the example screenshot above, Crowdstrike reported:
  • 🔼🔼🔼 revenue increased in all of the past three quarters
  • 🔺🔻🔻 In the past three quarters, earnings initially increased, then decreased for two quarters, and was negative in every quarter
  • 🔼🔼🔼 revenue increased in all of the past three years
  • 🔻🔺🔻 In the past three years, earnings decreased, increased, then decreased, and was negative for all three years

This report can only be run through the interactive bot. Example usage:

.AAPL

@botname AAPL

Company Profile

This report can only be run through the interactive bot. Example usage:

.profile [symbol]

@botname profile [symbol]

Trades

trades.py sends recent Sharesight trades to your configured chat services. By default, this report searches Sharesight for trades from the current day only. You can override this with past_days in the .env file, or by providing a number as argument when triggered through the chat bot.

Without persistent storage, it is recommended to leave past_days = 0 to avoid duplicate trade notifications. In this case, the cron that triggers the report must do so exactly once per day, after market close.

With persistent storage, it is recommended to set past_days = 30. This is useful if Sharesight imports trades with past dates for any reason. Note that the initial run will notify on all historical trades for the past_days period. It is recommended to set the cron frequency to 20 minutes.

past_days = 30

Cron trigger:

trades.py

Interactive trigger:

.trades [period]

@botname trades [period]

You can also specify a portfolio name to get trades for just that portfolio:

.trades [period] [portfolio]

@botname trades [period] [portfolio]

Price alerts

prices.py sends premarket, midsession, and interday price movements if any holdings moved over a percentage threshold. This data is sourced from Yahoo! Finance and Sharesight. The default threshold is 10% but you can change it by setting price_percent in the .env file, or by providing a number as argument when triggered through the chat bot. Decimal fractions are accepted.

If demote_volatile is true, stocks with a market cap under 150M, or a beta of > 1.5 and a market cap of under 1B will have their alert threshold (price_percent) doubled.

Config example:

price_percent = 9.4
demote_volatile = True

Cron trigger:

./price.py [premarket|midsession|interday|close|days (int)]

When scheduled (Cron), the mode must be passed as an execution argument:

• If premarket is passed, it only reports on pre/post market price movements.
• If midsession is passed, it only reports for markets currently in session. This is intended to run from Cron to provide mid-session alerts for big price movements of your holdings. For example, it could be run twice per day 12 hours apart, to capture markets in different timezones.
• If interday is passed, it reports current price against the previous market close.
• If close is passed, it reports daily change in stocks whose market closed in the past 3 hours.
• If an integer [i] is passed, it reports current price against [i] days ago.

Interactive trigger (pre-market):

.premarket [percent]

@botname premarket [percent]

Interactive trigger (mid-session):

.midsession [percent]

@botname midsession [percent]

Interactive trigger (interday).

.price [percent]

@botname [percent]

For interday, can also specify the time period.

.price [percent] [period]

Or a specific stock. In this case, the output includes a graph.

.price [symbol] [period]

Portfolio performance

performance.py reports on Sharesight portfolio performance over a given period. The default is set by past_days in the .env file, which can be overridden at runtime by specifying an integer as the argument.

Config example:

past_days = 28

Cron trigger:

./performance.py 7

Interactive trigger:

.performance [period]

@botname performance [period]

Interactively, you can also specify a portfolio:

.performance [period] [[portfolio]

@botname performance [period] [portfolio]

Events calendar

cal.py sends upcoming earnings and ex-dividend date alerts. The data is sourced from Yahoo! Finance. It reports on events up to future_days into the future. This is set in the .env file for triggering the report from Cron, or can be specified as an argument when triggered through the chat bot.

Finance Explainer: When a company releases its quarterly earnings report, the stock price may undergo a signficant positive or negative movement, depending on whether the company beat or missed market expectations. You may wish to hold off buying more of this stock until after its earnings report, unless you think the stock will beat market expectations.

Finance Explainer: When a stock goes ex-dividend, the share price typically drops by the amount of the dividend paid. If you buy right before the ex-dividend date, you can expect an unrealised capital loss, plus a tax obligation for the dividend. Thus, you may wish to wait for the ex-dividend date before buying more of this stock.

future_days = 7

Cron execution:

./cal.py [earnings|ex-dividend]

Interactive trigger (earnings):

.earnings [period]

@botname earnings [period]

Interactive trigger (ex-dividend):

.dividend [period]

@botname dividend [period]

Highly shorted stock warnings

shorts.py sends short interest warnings. The data is sourced from Yahoo Finance and Shortman (AU). shorts_percent defines the alert threshold for the percentage of a stock's float shorted. This can be specified in the .env file for running the report from Cron, or as an argument when triggered through the chat bot.

Finance Explainer: A high short ratio indicates a stock is exposed to high risks, such as potential banktrupcy. It may also incentivise negative news articles which harm the stock price. If the market is wrong, however, risk tolerant investors may receive windfall gains. This report is intended to alert you to an above-average risk, and prompt you to investigate this stock more closely.

shorts_percent = 15

Interactive trigger:

.shorts [percent]

@botname shorts [percent]

pe / forward pe / peg

List securities by P/E ratio, forward P/E ratio, or PEG ratio

Interactive trigger:

.pe [SYMBOL]
.fpe [SYMBOL|bottom|negative]
.peg [SYMBOL|bottom|negative]

@botname pe [SYMBOL|top|bottom]
@botname fpe [SYMBOL|top|bottom|negative]
@botname peg [SYMBOL|top|bottom|negative]

History

Shows the long term performance of a security. It creates a 5Y chart, and then lists performance for several periods (5Y 3Y 1Y 6M 3M 1M 1W 1D)

Interactive trigger:

.history <symbol>

@botname history <symbol>

Buy

Show top recommended buys by Yahoo Finance analysts rating

Interactive trigger:

.buy

@botname buy

Sell

Show top recommended sells by Yahoo Finance analysts rating

Interactive trigger:

.sell

@botname sell

Marketcap

List the top (or bottom) stocks by market cap

Interactive trigger:

.marketcap [top|bottom]

@botname marketcap [top|bottom]

Consensus rating changes

Notifies when Yahoo Finance analyst rating consensus changes category (e.g. from hold to buy). Designed to run from Cron. Requires persistent storage.

./rating.py

Watchlist

Tracks additional securities which are not in your Sharesight holdings. The various reports will include wishlist items as if they are holdings.

It is stored in var/cache/finbot_watchlist.json by default. It is a JSON array literal containing Yahoo symbols. Example:

["2454.TW", "3217.TWO", "ASO.AX", "STEM"]

When the bot is run interactively, the watchlist is dynamic and can be edited by members of a chat group. In this case, there is no need to create/edit the file manually.

Interactive trigger:

.watchlist

.watchlist [add|del] [symbol]

@botname watchlist

@botname watchlist [add|del] [symbol]

Installation (Linux)

sudo $(which apt dnf yum) install git python3-pip

git clone https://github.com/robdevops/finbot.git ~/finbot

cd ~/finbot

pip3 install -r requirements.txt

Setup

Configuration is set by the .env file in the main directory. Example:

vi ~/finbot/.env

Sharesight

• Email Sharesight support to get an API key and add the access details to the .env file. Example:

sharesight_code = '01d692d4de7 samplefc64bc2e2f01d692d4de72986ea808f6e99813f'
sharesight_client_id = '01d692d4de7 samplefc64bc2e2f01d692d4de72986ea808f6e99813f'
sharesight_client_secret = '01d692d4de7 samplefc64bc2e2f01d692d4de72986ea808f6e99813f'

Discord

Discord support is currently only for trade notifications and scheduled reports. The Discord webhook can be provisioned in Discord under Server Settings > Integrations > Webhooks. Example:

discord_webhook = 'https://discord.com/api/webhooks/1009998000000000000/aaaaaaaaaaaaaaaaaaaaaaaaaa-bbbbbbbbbbbbbb'

Slack

A Slack webhook can be provisioned by creating a new app from scratch at https://api.slack.com/apps/ then navigating to Incoming Webhooks. Once the link is generated, add it to the .env file. Example:

slack_webhook = 'https://hooks.slack.com/services/XXXXXXXXXXX/YYYYYYYYYYY/AAAAAAAAmockupAAAAAAAAAAAA'

The webhook is only used for the trade notifications and scheduled reports. The interactive bot requires further configuration of this Slack app (see below).

Telegram

• Set up the bot by messaging BotFather.
• Add your bot to a group or channel.
• For security, run /setjoingroups and set to Disabled
• Optionally, make the bot a group admin (for interactive features).
• In the .env file, set telegramBotToken to the token BotFather gave you.
• In the .env file, set telegramChatID to the chat group or channel id.
  • For channels and supergroups, telegramChatID should be negative and 13 characters. Prepend -100 if necessary.
  • Be aware a group id can change if you edit group settings and it becomes a "supergroup". Currently, the bot does not automatically handle this.
• Example .env entry:

telegramBotToken = '0000000000:AAAAAAAAAAAAAAAAAAAAAAAAAA'
telegramChatID = -1001000000000

Portfolios

Portfolios are auto-discovered, including other people's portfolios which are shared to you. To exclude specific portfolios, add their IDs to exclude_portfolios in the .env file:

exclude_portfolios = "100003 100004"

Alternatively, you can include only specific portfolios:

include_portfolios = "100001 100002"

Caching

Many objects are cached for just under one day by default. Cache is controlled by the settings below. Trades IDs are stored on disk, but trade queries generally don't result in a cache hit for functional reasons.

cache = True
cache_seconds = 82800

Hyperlinks

These .env file options relate to hyperlinks:

• hyperlink, if set to False, disables all hyperlinks. Note: it can not disable syntax highlighting and false positives by chat clients that try to detect domains. Setting it to False can be useful for chat networks like Discord that can't easily disable link previews. Defaults to True.

hyperlink = True

• hyperlinkFooter appends hyperlinks to the company website and to a Google search for the ticker on simplywall.st for the symbol lookup report. Possible values are True or False. Defaults to False.

hyperlinkFooter = True

• hyperlinkProvider configures where to send ticker hyperlinks. Available values are google and yahoo. If set to google, it is only used where practical, as Google does not publish earnings date, ex-dividend date, premarket price, or short interest figures. Defaults to yahoo.

hyperlinkProvider = google

hyperlinkProvider = yahoo

Scheduling example

Recommended for a machine set to UTC:

# Every 20 minutes on weekdays
*/20 * * * Mon-Fri ~/finbot/trades.py &> /dev/null

# Mid-session
00 01 * * Mon-Fri ~/finbot/price.py midsession &> /dev/null # AU
15 17 * * Mon-Fri ~/finbot/price.py midsession &> /dev/null # US

# Daily
30  21 * * * ~/finbot/reminder.py > /dev/null

# Daily on weekdays
29  21 * * Mon-Fri ~/finbot/price.py interday &> /dev/null
10  11 * * Mon-Fri ~/finbot/price.py premarket &> /dev/null

# Weekly
28  21 * * Sat { cd ~/finbot/; ./cal.py earnings; ./cal.py ex-dividend; ./price.py 7; ./performance.py 7; ./rating.py ;} &> /dev/null

# Monthly
27  21 1 * * { cd ~/finbot/; ./shorts.py; ./price.py 28; ./performance.py 28 ;} &> /dev/null

The above can be installed with:

(crontab -l ; cat ~/finbot/crontab.txt)| crontab -

Interactive bot setup

The backend bot.py needs a frontend https server on a valid domain name with a valid x509 certifcate. It defaults to listening on http://127.0.0.1:5000/, which can be changed with ip and port in the .env file.

Example frontend https server config (nginx):

server {
	listen 8443 ssl;
	deny all;

	server_name         www.example.com;
	ssl_certificate     /etc/letsencrypt/live/www.example.com/fullchain.pem;
	ssl_certificate_key /etc/letsencrypt/live/www.example.com/privkey.pem;
	ssl_protocols       TLSv1 TLSv1.1 TLSv1.2;
	ssl_ciphers         HIGH:!aNULL:!MD5;

	location /slack {
		proxy_pass http://127.0.0.1:5000/slack;
                include /etc/nginx/aws_subnets;
                deny all;
	}
	location /telegram {
		proxy_pass http://127.0.0.1:5000/telegram;
		include /etc/nginx/telegram_subnets;
		deny all;
	}
}

Note: The utils folder contains scripts to generate /etc/nginx/aws_subnets and /etc/nginx/telegram_subnets. They can be run once, or placed in /etc/cron.daily/.

Integrating the interactive bot with chat networks

Telegram

• Message BotFather to create a bot user.
  • Set .env file telegramBotToken to the token given by BotFather.
  • Add your Telegram bot user to a group
    • Give the bot group admin access so it can read the group chat.
    • For security, message BotFather and disable /setjoingroups.
• Set .env telegramOutgoingWebhook to your web server (e.g. https://www.example.com:8443/telegram).
• Set telegramOutgoingToken to a password of your choosing. Telegram will use this to authenticate with finbot. With these options set, your bot will auto-subscribe your URL to events the bot sees, when you run bot.py.

Slack

Visit https://api.slack.com/apps/ to create a new Slack app from scratch (if you already created one for a finbot webhook, you can reuse that app).

• From Basic Information > Verification Token, copy Verification token into the .env file variable slackOutgoingToken. This allows Slack to authenticate with finbot.
• In the .env file, put your web server URL (e.g. https://www.example.com:8443/slack) into slackOutgoingWebhook
• Save the .env file and (re)start bot.py
• Under Event Subscriptions:
  • Go to Enable Events > On > Request URL and enter your web server URL (e.g. https://www.example.com:8443/slack). At this time, if the bot is reachable, it will auto respond to Slack's verification request.
  • Go down to Subscribe to bot events and add event app_mention for the bot to see @botname mentions.
    • Alternatively, you can subscribe to message.channels if you want your bot to see everything and respond to . commands. To avoid duplicate responses, don't subscribe to app_mention and message.channels at the same time.
    • If you want to DM the bot:
      • Subscribe to message.im
      • Check the box App Home > Allow users to send Slash commands and messages from the messages tab
  • Save Changes
• Under OAuth & Permissions:
  • Scroll down to Scopes > Bot Token Scopes, and add: chat:write and files:write.
  • Scroll up to OAuth Tokens for Your Workspace:
    • If Bot User OAuth Token is not visible, hit Install to Workspace > Allow
    • Copy Bot User OAuth Token into .env file slackBotToken. This allows finbot to authenticate with Slack so it can send replies.
    • Restart bot.py

Daemonize (systemd)

finbot.service can take care of keeping bot.py running in the background and starting it on boot. Copy finbot.service to /etc/systemd/system/, edit it to set the User and ExecStart, then enable and start it:

sudo cp -v finbot.service /etc/systemd/system/

sudo sed -i 's/CHANGEME/YOUR LINUX USERNAME/' /etc/systemd/system/finbot.service

sudo systemctl daemon-reload

sudo systemctl enable finbot --now

You can now monitor the bot's stderr with :

journalctl -fu finbot

Limitations

• Discord shows garbage link previews from Sharesight. Set hyperlink = False, or disable this for each Discord client under Settings > Text & Images > Embeds and link previews (must be repeated for each participating Discord client).

Suggestions

• Is my code is doing something the hard way?
• Something important is missing from this README?

Log an issue!