Val Town Code SearchReturn to Val Town

9[![](https://stevekrouse-button.express.val.run/Install)](https://www.val.town/v/stevekrouse/blob_admin_app/fork)
10
11It uses [basic authentication](https://www.val.town/v/pomdtr/basicAuth) with your [Val Town API Token](https://www.val.town/settings/api) as the password (leave the username field blank).
12
13# TODO

8
91. Click `Fork`
102. Change `location` (Line 4) to describe your location. It accepts fairly flexible English descriptions which it turns into locations via [nominatim's geocoder API](https://www.val.town/v/stevekrouse/nominatimSearch).
113. Click `Run`
12

13Change the `query` variable for what you want to get notified for.
14
15You can use [Twitter's search operators](https://developer.twitter.com/en/docs/twitter-api/v1/rules-and-filtering/search-operators) to customize your query, for some collection of keywords, filtering out others, and much more!
16
17## 3. Notification

6* Fork this val to your own profile.
7* Make a folder for the temporary vals that get generated, take the ID from the URL, and put it in `tempValsParentFolderId`.
8* If you want to use OpenAI models you need to set the `OPENAI_API_KEY` [env var](https://www.val.town/settings/environment-variables).
9* If you want to use Anthropic models you need to set the `ANTHROPIC_API_KEY` [env var](https://www.val.town/settings/environment-variables).
10* Create a [Val Town API token](https://www.val.town/settings/api), open the browser preview of this val, and use the API token as the password to log in.
11
12<img width=500 src="https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/7077d1b5-1fa7-4a9b-4b93-f8d01d3e4f00/public"/>

6* Fork this val to your own profile.
7* Make a folder for the temporary vals that get generated, take the ID from the URL, and put it in `tempValsParentFolderId`.
8* If you want to use OpenAI models you need to set the `OPENAI_API_KEY` [env var](https://www.val.town/settings/environment-variables).
9* If you want to use Anthropic models you need to set the `ANTHROPIC_API_KEY` [env var](https://www.val.town/settings/environment-variables).
10* Create a [Val Town API token](https://www.val.town/settings/api), open the browser preview of this val, and use the API token as the password to log in.
11
12<img width=500 src="https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/7077d1b5-1fa7-4a9b-4b93-f8d01d3e4f00/public"/>

6* Fork this val to your own profile.
7* Make a folder for the temporary vals that get generated, take the ID from the URL, and put it in `tempValsParentFolderId`.
8* If you want to use OpenAI models you need to set the `OPENAI_API_KEY` [env var](https://www.val.town/settings/environment-variables).
9* If you want to use Anthropic models you need to set the `ANTHROPIC_API_KEY` [env var](https://www.val.town/settings/environment-variables).
10* Create a [Val Town API token](https://www.val.town/settings/api), open the browser preview of this val, and use the API token as the password to log in.
11
12<img width=500 src="https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/7077d1b5-1fa7-4a9b-4b93-f8d01d3e4f00/public"/>

1* Fork this val to your own profile.
2* Create a [Val Town API token](https://www.val.town/settings/api), open the browser preview of this val, and use the API token as the password to log in.
3

113. 🤣🤣🤣🤣
12
13## API
14
15This val uses the [icanhazdadjoke API](https://icanhazdadjoke.com/api). You can find [more docs here](https://github.com/15Dkatz/official_joke_api), such as how to [filter by type](https://github.com/15Dkatz/official_joke_api?tab=readme-ov-file#grab-jokes-by-type).

3
4export async function dailyDadJoke() {
5  let { setup, punchline } = await fetchJSON("https://official-joke-api.appspot.com/random_joke");
6  return email({
7    text: punchline,

1/**
2 * This code creates a search engine prototype with autocomplete functionality using the Cerebras LLM API.
3 * It uses React for the frontend and the Cerebras API for generating autocomplete suggestions.
4 * The suggestions are cached in the browser to reduce API calls.
5 * It implements a two-step LLM process: first to get initial suggestions, then to filter them for sensibility and ethics.
6 * If the second LLM call fails, it displays "Failed to fetch" instead of showing results.
104  if (request.method === "POST" && new URL(request.url).pathname === "/suggestions") {
105    const { query } = await request.json();
106    const apiKey = Deno.env.get("CEREBRAS_API_KEY");
107    const initialSuggestions: string[] = [];
108
109    if (!apiKey) {
110      return new Response(JSON.stringify({ error: "API key not found" }), {
111        status: 500,
112        headers: { "Content-Type": "application/json" },
115
116    try {
117      const response = await fetch("https://api.cerebras.ai/v1/chat/completions", {
118        method: "POST",
119        headers: {
120          "Content-Type": "application/json",
121          "Authorization": `Bearer ${apiKey}`,
122        },
123        body: JSON.stringify({
137
138      if (!response.ok) {
139        throw new Error("Failed to fetch from Cerebras API");
140      }
141
146
147      // Second LLM call to filter suggestions
148      const filterResponse = await fetch("https://api.cerebras.ai/v1/chat/completions", {
149        method: "POST",
150        headers: {
151          "Content-Type": "application/json",
152          "Authorization": `Bearer ${apiKey}`,
153        },
154        body: JSON.stringify({
189      });
190    } catch (error) {
191      console.error("Error calling Cerebras API:", error);
192      return new Response(JSON.stringify({ error: "Failed to fetch" }), {
193        status: 500,