Search

  handler
- Environment variables via `Deno.env.get()`
- Val Town std libraries used: `email` (forwarding), `openai` (LLM fallback)
## Code Standards
- **Geocoder**: US Census Bureau (`geocoding.geo.census.gov`) — free, no API
  key, handles DC intersections
- **LLM fallback**: OpenAI gpt-4o-mini via Val Town `std/openai` for locations
  that resist deterministic parsing
- **Email format**: Location is always in

  intersections well)
- **LLM fallback**: When deterministic parsing can't produce a geocodable
  address, OpenAI gpt-4o-mini rewrites the location string before a second
  geocode attempt
- **Email parsing**: Location is extracted from the pattern

_2 or _3) to create a fresh table.
### OpenAI
```ts
import { OpenAI } from "https://esm.town/v/std/openai";
const openai = new OpenAI();
const completion = await openai.chat.completions.create({
  messages: [
    { role: "user", content: "Say hello in a creative way" },

async function geocodeWithLLM(rawLocation: string): Promise<GeoResult | null> {
  const { OpenAI } = await import("https://esm.town/v/std/openai");
  const openai = new OpenAI();
  const completion = await openai.chat.completions.create({
    messages: [
      {

Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.
### OpenAI
```ts
import { OpenAI } from "https://esm.town/v/std/openai";
const openai = new OpenAI();
const completion = await openai.chat.completions.create({
  messages: [
    { role: "user", content: "Say hello in a creative way" },

import { IncomingWebhook } from "npm:@slack/webhook";
import { Octokit } from "npm:octokit";
import { OpenAI } from "npm:openai";
// Environment variables
  commits: any[],
): Promise<string> {
  const openai = new OpenAI();
  // Extract commit messages
  try {
    console.log("prompt", prompt);
    const completion = await openai.chat.completions.create({
      messages: [{ role: "user", content: prompt }],
      model: "gpt-5.2-2025-12-11",

// Vibe Engine - Synthesizes prompt stack into OpenAI calls

import { OpenAI } from "https://esm.town/v/std/openai";
import type { PlayedCard, PromptResult } from "../../shared/types.ts";

  const prompt = buildPrompt(config, userQuery);
  
  const openai = new OpenAI();
  
  const messages: any[] = [
  }
  
  const completion = await openai.chat.completions.create({
    model: "gpt-4o-mini",
    messages,

});

// API: Generate vibe (run OpenAI)
app.post("/api/session/:sessionId/generate", async c => {
  const sessionId = c.req.param("sessionId");

### Generation fails
Check that OpenAI integration is working in Val Town console.

## Tips for Fun Sessions

- Val Town account
- Val Town MCP tools installed
- OpenAI API key (automatically provided by Val Town)

## Deployment Steps
## Environment Variables

OpenAI API key is automatically provided by Val Town. No manual setup needed!

If you want to use your own API key:

```bash
mcp_valtown_add_env_var --val "<your-username>/vibe-draft" --key "OPENAI_API_KEY" --value "sk-..." --description "OpenAI API key"
```

Check that `shared/cards.ts` is uploaded correctly and importing properly.

### OpenAI Errors
Verify OpenAI API key is set (should be automatic). Check Val Town console for API errors.

### Database Errors