XML parsing in AWS Lambda function

Actually I just tested this and you can actually use xml2js straight out of the box because...

https://github.com/aws/aws-sdk-js/blob/master/lib/xml/node_parser.js

That's what the AWS JS SDK uses. Sample Lambda code use to test this, completely using the Lambda online editor and running test data against it:

'use strict';

var xml2js = require('xml2js');

console.log('Loading function');

var options = {  // options passed to xml2js parser
  explicitCharkey: false, // undocumented
  trim: false,            // trim the leading/trailing whitespace from text nodes
  normalize: false,       // trim interior whitespace inside text nodes
  explicitRoot: false,    // return the root node in the resulting object?
  emptyTag: null,         // the default value for empty nodes
  explicitArray: true,    // always put child nodes in an array
  ignoreAttrs: false,     // ignore attributes, only create text nodes
  mergeAttrs: false,      // merge attributes and child elements
  validator: null         // a callable validator
};

exports.handler = (event, context, callback) => {
    var parser = new xml2js.Parser(options);
    //console.log('Received event:', JSON.stringify(event, null, 2));
    console.log('value1 =', event.key1);
    console.log('value2 =', event.key2);
    console.log('value3 =', event.key3);
    callback(null, event.key1);  // Echo back the first key value
    //callback('Something went wrong');
};

That said if you want to avoid that route you're going to have to go the standard package install route.