You are an advanced assistant specialized in generating Val Town code.

• Ask clarifying questions when requirements are ambiguous
• Provide complete, functional solutions rather than skeleton implementations
• Test your logic against edge cases before presenting the final solution
• Ensure all code follows Val Town's specific platform requirements
• If a section of code that you're working on is getting too complex, consider refactoring it into subcomponents

• Generate code in TypeScript or TSX
• Add appropriate TypeScript types and interfaces for all data structures
• Prefer official SDKs or libraries than writing API calls directly
• Ask the user to supply API or library documentation if you are at all unsure about it
• Never bake in secrets into the code - always use environment variables
• Include comments explaining complex logic (avoid commenting obvious operations)
• Follow modern ES6+ conventions and functional programming practices if possible

• Create web APIs and endpoints
• Handle HTTP requests and responses
• Example structure:

Create val
export default async function (req: Request) {
  return new Response("Hello World");
}

Files that are HTTP triggers have http in their name like foobar.http.tsx

• Run on a schedule
• Use cron expressions for timing
• Example structure:

Create val
export default async function () {
  // Scheduled task code
}

Files that are Cron triggers have cron in their name like foobar.cron.tsx

• Process incoming emails
• Handle email-based workflows
• Example structure:

Create val
export default async function (email: Email) {
  // Process email
}

Files that are Email triggers have email in their name like foobar.email.tsx

Val Town provides several hosted services and utility functions.

Create val
import { blob } from "https://esm.town/v/std/blob";
await blob.setJSON("myKey", { hello: "world" });
let blobDemo = await blob.getJSON("myKey");
let appKeys = await blob.list("app_");
await blob.delete("myKey");

Create val
import { sqlite } from "https://esm.town/v/stevekrouse/sqlite";
const TABLE_NAME = 'todo_app_users_2';
// Create table - do this before usage and change table name when modifying schema
await sqlite.execute(`CREATE TABLE IF NOT EXISTS ${TABLE_NAME} (
  id INTEGER PRIMARY KEY AUTOINCREMENT,
  name TEXT NOT NULL
)`);
// Query data
const result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} WHERE id = ?`, [1]);

Note: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.

Create val
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" },
  ],
  model: "gpt-4o-mini",
  max_tokens: 30,
});

Create val
import { email } from "https://esm.town/v/std/email";
// By default emails the owner of the val
await email({ 
  subject: "Hi",  
  text: "Hi", 
  html: "<h1>Hi</h1>"
});

Val Town provides several utility functions to help with common project tasks.

Always import utilities with version pins to avoid breaking changes:

Create val
import { parseProject, readFile, serveFile } from "https://esm.town/v/std/utils@85-main/index.ts";

For example, in Hono:

Create val
// serve all files in frontend/ and shared/
app.get("/frontend/*", c => serveFile(c.req.path, import.meta.url));
app.get("/shared/*", c => serveFile(c.req.path, import.meta.url));

Create val
// Read a file from the project
const fileContent = await readFile("/frontend/index.html", import.meta.url);

Create val
const files = await listFiles(import.meta.url);

This is useful for including info for linking back to a val, ie in "view source" urls:

Create val
const projectVal = parseProject(import.meta.url);
console.log(projectVal.username); // Owner of the project
console.log(projectVal.name);     // Project name
console.log(projectVal.version);  // Version number
console.log(projectVal.branch);   // Branch name
console.log(projectVal.links.self.project); // URL to the project page

However, it's extremely importing to note that parseProject and other Standard Library utilities ONLY RUN ON THE SERVER. If you need access to this data on the client, run it in the server and pass it to the client by splicing it into the HTML page or by making an API request for it.

• Redirects: Use return new Response(null, { status: 302, headers: { Location: "/place/to/redirect" }}) instead of Response.redirect which is broken
• Images: Avoid external images or base64 images. Use emojis, unicode symbols, or icon fonts/libraries instead
• AI Image: To inline generate an AI image use: <img src="https://maxm-imggenurl.web.val.run/the-description-of-your-image" />
• Storage: DO NOT use the Deno KV module for storage
• Browser APIs: DO NOT use the alert(), prompt(), or confirm() methods
• Weather Data: Use open-meteo for weather data (doesn't require API keys) unless otherwise specified
• View Source: Add a view source link by importing & using import.meta.url.replace("ems.sh", "val.town)" (or passing this data to the client) and include target="_top" attribute
• Error Debugging: Add <script src="https://esm.town/v/std/catch"></script> to HTML to capture client-side errors
• Error Handling: Only use try...catch when there's a clear local resolution; Avoid catches that merely log or return 500s. Let errors bubble up with full context
• Environment Variables: Use Deno.env.get('keyname') when you need to, but generally prefer APIs that don't require keys
• Imports: Use https://esm.sh for npm and Deno dependencies to ensure compatibility on server and browser
• Storage Strategy: Only use backend storage if explicitly required; prefer simple static client-side sites
• React Configuration: When using React libraries, pin versions with ?deps=react@18.2.0,react-dom@18.2.0 and start the file with /** @jsxImportSource https://esm.sh/react@18.2.0 */
• Ensure all React dependencies and sub-dependencies are pinned to the same version
• Styling: Default to using TailwindCSS via <script src="https://cdn.twind.style" crossorigin></script> unless otherwise specified

├── backend/
│   ├── database/
│   │   ├── migrations.ts    # Schema definitions
│   │   ├── queries.ts       # DB query functions
│   │   └── README.md
│   └── routes/              # Route modules
│       ├── [route].ts
│       └── static.ts        # Static file serving
│   ├── index.ts             # Main entry point
│   └── README.md
├── frontend/
│   ├── components/
│   │   ├── App.tsx
│   │   └── [Component].tsx
│   ├── favicon.svg
│   ├── index.html           # Main HTML template
│   ├── index.tsx            # Frontend JS entry point
│   ├── README.md
│   └── style.css
├── README.md
└── shared/
    ├── README.md
    └── utils.ts             # Shared types and functions

• Hono is the recommended API framework
• Main entry point should be backend/index.ts
• Static asset serving: Use the utility functions to read and serve project files:

  Create val
  import { readFile, serveFile } from "https://esm.town/v/std/utils@85-main/index.ts";
  
  // serve all files in frontend/ and shared/
  app.get("/frontend/*", c => serveFile(c.req.path, import.meta.url));
  app.get("/shared/*", c => serveFile(c.req.path, import.meta.url));
  
  // For index.html, often you'll want to bootstrap with initial data
  app.get("/", async c => {
    let html = await readFile("/frontend/index.html", import.meta.url);
    
    // Inject data to avoid extra round-trips
    const initialData = await fetchInitialData();
    const dataScript = `<script>
      window.__INITIAL_DATA__ = ${JSON.stringify(initialData)};
    </script>`;
    
    html = html.replace("</head>", `${dataScript}</head>`);
    return c.html(html);
  });

• Create RESTful API routes for CRUD operations
• Always include this snippet at the top-level Hono app to re-throwing errors to see full stack traces:

  Create val
  // Unwrap Hono errors to see original error details
  app.onError((err, c) => {
    throw err;
  });

• Run migrations on startup or comment out for performance
• Change table names when modifying schemas rather than altering
• Export clear query functions with proper TypeScript typing

1. Environment Limitations:

   • Val Town runs on Deno in a serverless context, not Node.js
   • Code in shared/ must work in both frontend and backend environments
   • Cannot use Deno keyword in shared code
   • Use https://esm.sh for imports that work in both environments

2. SQLite Peculiarities:

   • Limited support for ALTER TABLE operations
   • Create new tables with updated schemas and copy data when needed
   • Always run table creation before querying

3. React Configuration:

   • All React dependencies must be pinned to 18.2.0
   • Always include @jsxImportSource https://esm.sh/react@18.2.0 at the top of React files
   • Rendering issues often come from mismatched React versions

4. File Handling:

   • Val Town only supports text files, not binary
   • Use the provided utilities to read files across branches and forks
   • For files in the project, use readFile helpers

5. API Design:

   • fetch handler is the entry point for HTTP vals
   • Run the Hono app with export default app.fetch // This is the entry point for HTTP vals

• Use the relevant log level based on what you're logging.
• Import https://www.val.town/x/cricks_unmixed4u/logger/code/logger/main.tsx and use logInfo, logError or logDebug.

Updated pseudo-PRD with “today-only” constraint.

---

Activity Log MVP – “Today-only form-first tracking”

---

You log activities with a fixed schema (date, time, mood with score, overstimulation level, “must” activity, pleasant/relaxing activity). You want a tiny app that:

• Always works on “today”
• Makes time-range entry extremely easy for the next activity block
• Lets you refine the structure over time.

---

1. Fast capture of activity rows for today

   • No date picker, no navigation.
   • The app is implicitly “today’s log”.

2. Respect the existing schema

   • For each row, capture:

     • date (implicit: today)
     • time range
     • mood (with numeric rating)
     • overstimulation level
     • must activity
     • pleasant/relaxing activity

3. Support both free-text and structured entry

   • Free-text description → optional LLM assist to fill fields.
   • Full manual control over all fields before saving.

4. Simple persistence + today view

   • Store all rows in a simple store.
   • Show all entries for today in a table that mirrors the activities form.

5. Be easy to extend later

   • Clear separation between:

     • Storage
     • LLM parsing
     • UI

---

• User: you, logging only for today.

• Usage pattern:

  • Multiple short entry bursts during today.
  • Some retrospective entries within the same day (e.g. “what happened from 08:30–09:15?”).

Out of scope for MVP: viewing or editing past days.

---

• App loads with:

  • A “New entry for today” panel.
  • A table showing all entries already recorded for today.

Date is implied and not editable.

In the “New entry” panel:

1. Time range selection (always for today)

The app knows the latest endTime of today’s entries (if any).

Time range modes:

• Mode A – From last end → now

  • Available if at least one previous entry exists today.
  • Default mode.
  • Start = last entry’s end time (read-only by default).
  • End = “now” (auto-filled, but can be overridden).
  • One click to accept this range.

• Mode B – From custom start → now

  • For retroactive blocks earlier today.
  • User inputs startTime (e.g. 08:30).
  • End is auto-set to “now” (editable if needed).

• Mode C – From custom start → custom end

  • For explicit blocks within today (e.g. 09:15–10:00).
  • User inputs both startTime and endTime.

First entry of the day:

• No “last end”; default to Mode B or C with empty time fields.
• You fill the first block explicitly.

2. Content entry

Two ways to fill the content columns:

• Structured fields

  • Mood with score (single text field, must contain “(x/10)”).
  • Overstimulation level (e.g. free text or low/medium/high).
  • Must activity (text, "-" allowed).
  • Pleasant/relaxing activity (text, "-" allowed).

3. Save

Before saving:

• Validate that startTime < endTime.
• Validate that the mood includes a numeric score pattern like (6/10).

On success:

• Store the entry with date = today.
• Append it to today’s table.
• Update “last end” to this entry’s endTime.

• A table below the input panel:

Header:

Time | Mood (with rating) | Overstimulation level | Must activity | Pleasant/relaxing activity

• Each row shows:

  • Time as HH:mm–HH:mm
  • The four content columns.

No navigation to other dates; only today is visible.

---

Per entry:

• id (string or numeric)
• date (ISO date string, always today in this MVP)
• startTime ("HH:mm")
• endTime ("HH:mm")
• moodWithScore (string, contains “(n/10)”)
• overstimulationLevel (string)
• mustActivity (string, "-" allowed)
• pleasantActivity (string, "-" allowed)
• rawInput (optional string; free-text source, if used)
• createdAt (timestamp)

---

1. HTTP endpoints

• GET /

  • Renders HTML for:

    • New-entry panel (today-only).
    • Today’s table.

• GET /entries/today

  • Returns today’s entries as JSON.

• POST /parse

  • Input: { rawText, startTime?, endTime? } (today is implied).
  • Output: suggested moodWithScore, overstimulationLevel, mustActivity, pleasantActivity.

• POST /entries

  • Input: full entry payload (except date which is set to today on the server).

2. Storage

• vltable / JSON keyed by date and time.
• Query: “entries where date = today, sorted by startTime”.

3. LLM helperG

• Function parseActivityEntry(rawText, startTime, endTime):

  • Uses activities-form prompt.
  • Returns the four content fields in a strict schema.

---

• No date navigation or historical views.
• No multi-user support.
• No analytics or charts.
• No complex editing or versioning (append-only for today in MVP).

Let's add the UI components with data backed in local storage so we don't have to write backend endpoints.

For now, let's not style the app beyond using tailwind to have some paddings, font and form styling.

Instead of accepting a 1-10 range, let's accept a 1-5 range and help user just click/tap on a number for faster data entry. Don't have a default selection. Focus on mobile user input experience with a touch screen.

Also for overstimulation.

1. Define a versioned rating model with Zod

   • Create a Rating schema that represents “an ordinal rating on a scale”, with:

     • A dimension field: "mood" or "overstimulation".
     • A scale definition: min, max, step.
     • A value that must lie on that scale.
     • Optional ui metadata (e.g. style: buttons, slider, faces).
     • Optional note (for mood text like “calm, tired”).

   • Wrap this in a versioned structure (for now version: 1), so you can later add version: 2 without breaking old data.

2. Define an Entry schema with Zod

   • Create a Zod schema for an activity entry that includes:

     • date, startTime, endTime as validated strings (YYYY-MM-DD, HH:mm).
     • Canonical numeric scores: moodScore (1–5), overstimulationScore (1–3), nullable.
     • Optional moodRating and overstimulationRating as your versioned rating objects.
     • mustActivity, pleasantActivity, createdAt, id.

   • Derive the TypeScript Entry type from this Zod schema, so the schema is the single source of truth.

3. Let the form work with rating objects, not just numbers

   • In the “new entry” UI:

     • Keep the tap UI as simple 1–5 and 1–3 buttons for now.
     • Internally, when a button is tapped, construct a rating object (scale + value + dimension = "mood" or "overstimulation").

   • Treat the numeric score as a derived field from the rating (rating.value), not the primary state.

4. Validate entries through Zod before saving

   • When the user taps “save”:

     • Build an Entry object that includes:

       • Time range, activities, createdAt, id.
       • moodRating and overstimulationRating.
       • moodScore and overstimulationScore copied from those ratings’ value.

     • Run this through the Zod entrySchema:

       • On success: store it (localStorage for now) and update in-memory state.
       • On failure: surface a simple error (for you, this is primarily to catch bugs).

5. Use Zod when loading from localStorage

   • When restoring data from localStorage:

     • Parse the JSON and validate it as “array of Entry” with Zod.

     • If something does not conform to the schema, either:

       • Drop those entries and log the error, or
       • Apply a small manual migration before validation (e.g. fill missing moodRating from moodScore).

6. Keep the versioning future-proof

   • Always put the version field inside the rating object.
   • Wrap your rating schema as a union over versions (for now just V1).
   • This lets you later introduce a different input style or scale (e.g. mood 0–100 slider) as version: 2 without touching existing data or frontend logic beyond extending the union.

Once this is in place, your frontend is already “rating-version aware,” and switching from localStorage to a database later is mostly about wiring these same Entry objects through an API rather than changing their structure.

Move your ratingSchema and entrySchema into a shared module (e.g. models/activity.ts) that both frontend and backend can import.

Backend should:

Use the same Entry and RatingPayload types (via z.infer).

Use the same Zod schemas for request validation and response shaping.

Goal: one source of truth for the data model.

Add columns for the canonical numeric scores:

mood_score (small integer, nullable).

overstimulation_score (small integer, nullable).

Add JSON columns for the versioned rating payloads:

mood_rating_json (JSON/JSONB, nullable).

overstimulation_rating_json (JSON/JSONB, nullable).

Keep existing time and activity fields as-is:

date, start_time, end_time, must_activity, pleasant_activity, created_at.

Goal: DB row can store both “fast to query” numbers and “rich, versioned” JSON.

Decide on a clear mapping:

date ↔ Entry.date (YYYY-MM-DD).

start_time / end_time ↔ startTime / endTime.

mood_score ↔ moodScore.

overstimulation_score ↔ overstimulationScore.

mood_rating_json ↔ moodRating.

overstimulation_rating_json ↔ overstimulationRating.

Implement small helpers:

dbRowToEntry(row): converts a DB row to a plain object and runs entrySchema.parse on it.

entryToDbParams(entry): picks out the DB fields for inserts.

Goal: all API handlers work in terms of Entry objects, not raw DB rows.

For POST /entries:

Define an “input schema” that represents what the client sends (no id, createdAt).

Use Zod to:

Validate and parse the incoming JSON into a partial Entry.

Fill id and createdAt on the backend.

Before inserting:

Confirm moodRating and overstimulationRating (if present) pass ratingSchema.

Ensure their value matches moodScore / overstimulationScore (or derive scores from value and ignore client-provided numbers).

Goal: backend never trusts raw JSON; everything passes through Zod.

Since we don't have authentication, for now we can simply store a user ID in the localstorage and use that.

For now let's just focus on storing the data, later we can also use it in the frontend.

Goal: when there are no entries for today yet, the app should start with a minimal, calm “How did you wake up?” flow, and only then show the regular timeline.

1. Detect “first entry of the day”

On app load: Compute today. Load today’s entries (from localStorage/DB). If todayEntries.length === 0, the initial UI state is: Show a dedicated “wake-up check-in” screen/card. If there are existing entries: Skip this; show the normal new-entry form + today timeline. Keep this decision in a small helper like shouldShowWakeCheckIn(todayEntries, now) so future time-based rules can plug in.

Make a DEFAULT_WAKE_BLOCK_MINUTES configuration that'll be used when converting wake-up time → time range.

1. Remove explicit time modes

   • Remove the “Time Range Mode” label and the three mode buttons.
   • Keep a single “Time Range” section with two fields: startTime, endTime.

2. Time fields

   • Both startTime and endTime are editable HH:mm values.
   • Render them as large, tap-friendly controls suitable for mobile.
   • Always show both fields on the “New Entry for Today” card.

3. Defaults when today has entries

   • Compute lastEndTime = last(todayEntries).endTime (todayEntries sorted chronologically).
   • On form init and after each successful save:
     • Set startTime = lastEndTime.
     • Set endTime = nowRoundedTo5Min().

4. Retro blocks (implicit modes via edits)

   • Do not expose any “mode” selector.
   • Interpret the user’s edits as follows:
     • If user keeps defaults → “from last end → now”.
     • If user edits only startTime → “custom start → now”.
     • If user edits both startTime and endTime → “custom start → custom end”.
   • Optional UI shortcuts (not required for logic):
     • When startTime is empty, show a clickable hint “Use last end (HH:MM)” that fills startTime.
     • When endTime is empty, show a clickable hint “Use now (HH:MM)” that fills endTime.

English title suggestion: use “Mood description” for stemming.

Tiny programming TODOs:

1. Naming decision

   • Use UI label: "Mood description".
   • Use field name in code: moodDescription.

2. Update shared models (frontend only for now)

   • Add moodDescription: string to the Entry Zod schema (required or optional – choose).
   • Regenerate Entry TypeScript type from the updated schema.
   • Ensure moodDescription is included in any Entry constructors/builders.

3. Update “New Entry for Today” UI

   • Add a text input under “Mood Score (1–5)” with label “Mood description”.
   • Bind this input to local state moodDescription.
   • On save, pass moodDescription into the Entry builder.

4. Wake-up check-in UI (first entry)

   • Add the same “Mood description” field to the wake-up card.
   • Bind to moodDescription state used for the first Entry.
   • Ensure wake-up Entry gets moodDescription.note`.

5. Save flow

   • Update the Entry object created on save to include moodDescription.
   • Run the updated entrySchema validation (it should now include moodDescription).
   • Confirm localStorage (or in-memory) entries now contain moodDescription.

6. Load flow

   • Update localStorage loading to read moodDescription from stored entries.
   • If older entries do not have moodDescription, default to "".
   • Re-save migrated entries only if needed (optional).

7. Display in “Today’s Log”

   • Add a “Mood description” column (or subline) in the log view.
   • Render moodDescription if present; otherwise hide or show “—”.

8. Backend

   • Add mood_description column to DB schema (text).
   • Map mood_description ↔ moodDescription in dbRowToEntry / entryToDbParams.

Add a small static module moodWords.ts (frontend only).

Define four buckets (2×2 model – energy × pleasantness):

highEnergyPleasant: string[]

highEnergyUnpleasant: string[]

lowEnergyPleasant: string[]

lowEnergyUnpleasant: string[]

Keep each list small for MVP (e.g. 6–10 words per bucket).

Ensure all words are simple, everyday language (no jargon).

Add a tiny “Help me describe this” link/button next to the “Mood description” input.

When pressed, open a lightweight helper UI:

Could be inline below the field, not a full-screen dialog (MVP).

Show a simple 2×2 layout, text-only, no graphics needed:

Row labels: High energy, Low energy.

Column labels: Unpleasant, Pleasant.

Under each quadrant label, display a few chips/buttons from the corresponding list:

Example layout:

High energy / Pleasant → chips: “excited”, “curious”, …

High energy / Unpleasant → chips: “angry”, “overwhelmed”, …

Low energy / Pleasant → chips: “calm”, “relieved”, …

Low energy / Unpleasant → chips: “tired”, “sad”, …

On tap of a word chip:

If mood description is empty → set it to that word.

If mood description already has content → append with a comma and space.

Do not auto-close the helper; allow multiple taps.

Add a small “Done” / “Close” link under the helper to hide it.

Add a small helper function to suggest a quadrant based on existing numeric scores:

Use moodScore (pleasant–unpleasant axis).

Use overstimulationScore as a crude energy proxy (low–high).

Use this only to:

Option A (simplest): visually emphasise one quadrant (e.g. show it first or bold header).

Option B (if simpler): do nothing for now; just keep helper independent of scores.

Do not make behavior dependent on this in v1; keep it purely as a hint or future hook.

Ensure mood description field already flows into:

Entry.moodDescription.

moodRating.note.

No additional persistence needed for the helper; it just writes into the existing field.

Keep the helper visually light:

No extra navigation.

Minimal text; mostly labels + chips.

Make all chips large, touch-friendly targets.