Skip to content

Latest commit

 

History

History
100 lines (69 loc) · 8.81 KB

README.md

File metadata and controls

100 lines (69 loc) · 8.81 KB
Raw

Trade Stocks with Alexa/Google Home

Ever read a news article that made you want to immediately buy or sell a stock? Getting to your portfolio via a web or mobile interface can be cumbersome and time-consuming, but now you can trade stocks within seconds using your voice assistant and the power of IoT!

Table of Contents

Requirements

  • Alpaca account (or any other trading platform with an accessible API)
  • AWS account (only the free tier is needed)
  • IFTTT account
  • Google Assistant (Alexa will also work)

Setting up Alpaca

  1. Sign up for an Alpaca account.

  2. If you'd like to trade with real money, click Individual Account and fill in all the required information. To do paper trading, no additional information is necessary. Simply click Go to Paper Account on the top-left. I'm using a Paper Trading account, but most of the steps should be similar across both. Step 2

  3. Head to the Overview Dashboard page, and on the right side you'll see a section titled Your API Keys. Click View to expand the section and then Generate New Key. Copy your API Key ID and Secret Key, as you won't be able to see it again unless you generate a new one. These will be used in the next section when creating the Lambda function. Step 3

Building an API to Run a Lambda Function with AWS

Creating the Lambda Function

  1. Sign up for an AWS free tier account, and then go to the AWS Lambda console page. You should see a Create function button. Step 1

  2. Click Create function, and give it a name. Set the Runtime to Python 3.8, and then click Create function.

  3. You should now be on the homepage for your newly-created Lambda function. Below the function name in the Designer section, click Layers, and then Add a layer. Lambda functions don't natively support the requests library for Python, so we're going to add it here. Step 3

  4. Under Choose a layer, click Specify an ARN and then enter arn:aws:lambda:us-east-2:770693421928:layer:Klayers-python38-requests:14. Replace us-east-2 with whichever AWS region you are currently using (you can find that by clicking the menu on the top-right corner, next to your name). Finally, click Add. Step 4

  5. Scroll down to the Function code section, and in the text editor, replace everything with this code. On line 44, replace ID and SECRET with the Alpaca API Key ID and Secret Key you copied when making your Alpaca account, keeping the quotes around them. Also comment out lines 48-50, unless you would like to create some kind of notification system every time you make a trade (won't be discussing that here). Finally, on the top-right of the Function code section, click Deploy. Step 5

  6. On the top-right of the Lambda function page, next to Throttle, Qualifiers, and Actions, click on Test. Set the event template to hello-world, and give the event a name. Replace the JSON body with the text in the image below, and then click Create. Finally, click the Test button again, and you should see a message that says Execution result: succeeded. Step 6-a Now if you go to your Alpaca account and check Paper Orders on the sidebar, you should see that an order has been placed for 5 shares of AMZN. Step 6-b

Exposing the Lambda Function with an API

  1. Now that you have a working Lambda function that trades stocks, the next step is to make it accessible from outside the AWS console. To do this, navigate to the AWS API Gateway homepage. From here, create a new REST API (NOT the private one). Step 1

  2. Under Create new API, select New API, and give the API a name. Make sure the Endpoint Type is set to Regional, and then click Create API.

  3. Click Actions, and then Create Method. A new dropdown box should appear under Resources. Click on it and select POST, and then click the checkmark. Step 3

  4. Click the green POST button under Resources, and then set the Integration type to Lambda Function. In the Lambda Function textbox, enter the name of the Lambda function you created earlier. Finally, click Save and OK. Step 4

  5. Under the Actions menu, click Deploy API. Set the Deployment stage to [New Stage] and give the stage a name, such as prod. Then click Deploy. Step 5

  6. You should now see a blue box with an invoke URL. Congrats! Your Lambda function is now accessible via a REST API. (Clicking the link won't do anything though, since it only has a POST method and requires some input.) Step 6

Using IFTTT to Connect a Voice Assistant with the API

  1. Create an IFTTT account, and then create a new applet. First we are going to build an applet to buy stocks, and then one to sell them. Next to If This, click Add, and search for Google Assistant. After selecting Google Assistant, select the trigger Say a phrase with both a number and a text ingredient. Step 1

  2. At this point you will have to connect your Google account to IFTTT, so click Connect and then give IFTTT the permissions it needs. (If you want to revoke permissions later, you can do that at any time on the Google Account permissions page.)

  3. Once connected, you should be on the page titled Complete trigger fields. Here, under What do you want to say? and What's another way to say it? (optional), you can add your ideal trigger phrase, using $ where the stock symbol will be said and # where the desired quantity will be said. I am using Buy # shares of $ and Get # shares of $. Finally, for What do you want the Assistant to say in response?, you can put something like Okay, getting # shares of $. Set the language to English, and then click Create trigger. Step 3

  4. Next, click the Add button next to Then That, and search for and select Webhooks. Click Make a web request, and click Connect. Now set the URL to your invoke URL that you saw earlier when creating the API on AWS, and set the method to POST. Under Content Type, select application/json, and set the Body to {% raw %}{"action":"buy","symbol":"{{TextField}}","quantity":"{{NumberField}}"}{% endraw %}. Finally, click Create action, and then Continue. Give the applet a name, or leave it as is, and then hit Finish. Step 4

  5. Repeat steps 1-4 to create another applet, but this time set the trigger phrases to something like Sell # shares of $, and set the action Body to {% raw %}{"action":"sell","symbol":"{{TextField}}","quantity":"{{NumberField}}"}{% endraw %}. Congrats! You're now ready to trade stocks on-the-fly with your voice assistant. Step 5

Usage

Use the phrases you wrote in your IFTTT triggers to make trades, and make sure to spell out the letters of the stock symbol to ensure that the voice assistant processes your input correctly (e.g. Hey Google, get 5 shares of A-M-Z-N). The trade should then show up in your Orders page on Alpaca.

<script src="https://cdn.signalfx.com/o11y-gdi-rum/latest/splunk-otel-web.js" crossorigin="anonymous"> </script> <script> SplunkRum.init({ beaconUrl: 'https://rum-ingest.us1.signalfx.com/v1/rum', rumAuth: 'JUeo4BsZeSWJ8QA7RKV2yA', app: 'personal-website' }); </script>