WEBVTT 00:00.240 --> 00:06.120 Okay, so I checked out this branch which is called project slash react search agent. 00:06.280 --> 00:12.000 And just a quick note if you want to check out the code for this section, check out this branch over 00:12.000 --> 00:12.520 here. 00:12.520 --> 00:18.280 And for technical reasons, I committed the code to this branch over here and not the branch on the 00:18.280 --> 00:19.320 bottom left here. 00:20.000 --> 00:23.480 Alright, let's start by initializing our Python project. 00:23.480 --> 00:24.800 Let's write UV init. 00:25.560 --> 00:28.520 And now let's go install the dependencies. 00:28.520 --> 00:30.160 Let's add link chain. 00:30.480 --> 00:32.520 We want to use link chain OpenAI. 00:33.040 --> 00:36.520 And we discussed those packages in previous videos. 00:37.040 --> 00:41.640 So now we want to add a new package which is called link Chain. 00:42.440 --> 00:49.680 And Tivoli is a third party which is going to help us connect our agent to the web and give the agent 00:49.720 --> 00:54.360 searching capabilities so it will act as a search engine for our agent. 00:54.360 --> 01:01.480 And it also has other very useful APIs like Tivoli Crawl, Tivoli Map, Tivoli Extract, which are very 01:01.480 --> 01:04.080 useful services for AI agents. 01:04.360 --> 01:10.060 And we'll be using those capabilities in direct section of the course, so we'll cover them as well. 01:10.260 --> 01:17.300 And by the way, today is actually the most popular choice for integrating a web search into an agent. 01:17.580 --> 01:25.100 And they're even featured in the official documentation as the default service for a search engine when 01:25.140 --> 01:26.540 implementing agents. 01:26.700 --> 01:29.020 So they really did go to example. 01:29.300 --> 01:33.700 And I think the reason for that is that they first of all have an excellent API. 01:33.740 --> 01:34.860 It scales very well. 01:34.900 --> 01:36.100 It's very easy to use. 01:36.340 --> 01:40.460 And I think they were actually the first ones to go and integrate this. 01:40.460 --> 01:44.380 So to connect an agent A with searching capabilities. 01:44.420 --> 01:49.140 Now it's also important to note that they have a very generous free tier. 01:49.340 --> 01:55.460 And you can see it's 1000 API requests per month, which is going to be more than enough for this course. 01:56.260 --> 01:56.540 All right. 01:56.540 --> 02:00.460 Let me just show you a quick demo of the capabilities that we can integrate with it. 02:01.180 --> 02:08.700 So in the API playground let me write a query like what are the latest anthropic models. 02:08.980 --> 02:12.560 And we can see in the response we get here the URL of the source. 02:12.760 --> 02:15.520 We also get the content you can see here. 02:15.560 --> 02:18.360 It's talking about Cloud Outpost 4.1. 02:18.760 --> 02:23.480 And in this search result it talks about Cloud Ops and site four. 02:24.600 --> 02:26.760 And the search can be much more advanced. 02:26.800 --> 02:30.640 We can have some special filtering and advanced capabilities. 02:30.640 --> 02:34.160 However in this section we're only going to stick with the basics. 02:34.200 --> 02:36.160 All right so that's enough about me. 02:36.280 --> 02:39.600 And I'm going to link to in the videos resources. 02:40.600 --> 02:41.200 Alrighty. 02:41.200 --> 02:42.720 Let's go back to the code here. 02:42.880 --> 02:47.920 And link chain Tavileh is the Tavileh link chain integration. 02:48.160 --> 02:54.520 And here the provider is Tavileh in contrast to open AI in the previous example. 02:54.760 --> 03:00.960 So a provider of link chain doesn't have to be a vendor which offers large language models. 03:00.960 --> 03:07.000 It can also be a vendor which offer services that are exposed via API's and clients. 03:07.000 --> 03:12.760 So in this case Tavileh is going to expose their services, their searching capabilities, their scraping 03:12.760 --> 03:19.500 capabilities, and the link chain integration here is going to make our lives very, very easy when 03:19.500 --> 03:25.740 we want to integrate those capabilities into our link chain link graph applications here, because the 03:26.220 --> 03:30.940 team they wrote now, this integration, they connected to all the link chain objects, which we're 03:30.940 --> 03:31.540 going to use. 03:31.540 --> 03:32.540 And I'll give you a quick hint. 03:32.540 --> 03:34.260 It's going to be link chain tools. 03:34.260 --> 03:37.500 And this is another example of a link chain provider. 03:37.820 --> 03:42.700 And I also forgot to add please also install Python. 03:42.700 --> 03:50.580 And this is the original SDK which we are going to also be using in order to show you something very 03:50.580 --> 03:51.500 very important. 03:52.260 --> 03:59.260 All right let's go and also install Python dot env to load environment variables like we did before. 03:59.540 --> 04:03.220 And let me also install black insert for some formatting. 04:03.460 --> 04:06.100 And uh yeah let's go and install everything. 04:06.100 --> 04:07.660 So let me go and click enter. 04:09.460 --> 04:13.660 We can check out the packages in the Pyproject.toml file over here. 04:13.660 --> 04:17.300 And if you want the exact packages we can check out the UV dot log. 04:17.660 --> 04:21.030 And let's go to our main.py file. 04:21.070 --> 04:22.190 The boilerplate code. 04:22.230 --> 04:23.110 Let's try it. 04:23.110 --> 04:25.830 And let's run this code and make sure that it's working. 04:26.470 --> 04:26.950 All right. 04:26.950 --> 04:28.310 So we can see it's working. 04:28.630 --> 04:31.910 And now let me show you what I have in my env file. 04:32.150 --> 04:38.790 So like in the previous section I have my open API key I have Lang Smith. 04:38.830 --> 04:40.270 Tracing equals true. 04:40.270 --> 04:45.910 So we want to trace everything and we want to check out the traces I have here my link Smith API key. 04:45.950 --> 04:51.910 Like previous sections I have my Lang Smith project which I changed in code search agent. 04:52.110 --> 04:58.830 And I also have here a API key which I got from the UI. 04:59.150 --> 05:01.390 All right, let me show you how you can get it. 05:01.390 --> 05:04.270 You simply need to go to Tavileh to log in. 05:07.630 --> 05:09.830 Let me now go and log with Google. 05:12.310 --> 05:14.550 And right here I have my environment. 05:14.550 --> 05:16.150 You can see all my API keys. 05:16.470 --> 05:18.190 Let me go and generate an API key. 05:18.230 --> 05:19.990 For example I'm using the free plan. 05:20.030 --> 05:20.550 By the way. 05:20.990 --> 05:27.250 Let me go and click here this plus button Now let's name the API key. 05:28.250 --> 05:29.650 Let's call it API key. 05:30.210 --> 05:32.530 I'm going to limit my monthly usage. 05:32.810 --> 05:35.050 And yeah let's hit create. 05:35.170 --> 05:38.690 And here you can simply go you can check out you can review the API key. 05:38.730 --> 05:39.970 You can copy it. 05:39.970 --> 05:43.650 And this is the value that you should put in your um dot env file. 05:44.450 --> 05:47.010 So let me go back now to the env file. 05:47.410 --> 05:54.010 And it's also important to note that the name of the environment variable needs to be underscore API 05:54.010 --> 05:59.130 key, because that's where link chain is going to be looking for the credentials to authenticate to 05:59.170 --> 06:00.450 the services. 06:00.450 --> 06:07.890 So it's working very similarly to the OpenAI API key here, where it's going to look for that API key 06:07.930 --> 06:09.410 in the environment variables. 06:09.650 --> 06:13.210 Alrighty, let's go now to our main.py file. 06:13.210 --> 06:15.530 And let's start with the imports. 06:15.810 --> 06:21.970 And first let's go and let's load all the environment variables from the env file like we did in the 06:21.970 --> 06:23.010 previous video. 06:23.010 --> 06:26.010 So let's import the load function. 06:26.010 --> 06:29.690 And I want to call it in the beginning of the file here.