Update stock-tracker extension

[vinayh]

Hello!

I've been using this extension regularly over the past few weeks and thought I should contribute back some minor fixes and (IMO) UX improvements. All were AI-assisted and human-verified.

Summary

• New: pre/post-market indicator (sunrise/moon icon) on rows showing extended-hours prices.
• New: ★ next to a search-result symbol when it's already in your favorites.
• New: last-updated timestamp shown next to the Favorites / Search Results section header.
• Fix: "Move Down in Favorites" was a no-op for the last item; both move actions now guard against missing symbols too.
• Fix: prices of 0 were rendering as the missing-value placeholder; negative values now scale with k/M/B/T suffixes.
• Robustness: refresh Yahoo Finance cookie/crumb proactively after 12 h instead of waiting for a request to fail.
• Robustness: only refresh cookie/crumb on 401/403 (was previously refreshing on any 4xx/5xx, including 404/5xx).
• Robustness: fix Set-Cookie parsing under native fetch (use getSetCookie()) and thread AbortSignal through cookie/crumb fetches.
• Robustness: cancel in-flight requests on view unmount.
• Robustness: recover gracefully from corrupted local-storage values (favorites + cached cookie/crumb).
• Internal: switch to native fetch, removing the punycode deprecation warning. Update @raycast/api, TypeScript, ESLint, and other dependencies; migrate to ESLint flat config.
• Internal: lift useFavorites / useStockInfo / useStockSearch to the top-level Command() for a single source of truth (eliminates duplicate LocalStorage loads on each search/clear toggle).

Test plan

• npm run build passes
• npm run lint passes (only the pre-existing @raycast/prefer-title-case warning, which is a false positive on "Move Up in Favorites")
• Verified in Raycast (npm run dev): favorites view, search, add/remove favorite, move up/down, detail panel, last-updated timestamp, market-state icons, ★ on already-favorited search results

[raycastbot]

Thank you for your first contribution! 🎉

🔔 @hmarr @ericchernuka @MeleMannen you might want to have a look.

You can use this guide to learn how to check out the Pull Request locally in order to test it.

📋 Quick checkout commands

BRANCH="stock-tracker-improvements"
FORK_URL="https://github.com/vinayh/raycast-extensions.git"
EXTENSION_NAME="stock-tracker"
REPO_NAME="raycast-extensions"

git clone -n --depth=1 --filter=tree:0 -b $BRANCH $FORK_URL
cd $REPO_NAME
git sparse-checkout set --no-cone "extensions/$EXTENSION_NAME"
git checkout
cd "extensions/$EXTENSION_NAME"
npm install && npm run dev

We're currently experiencing a high volume of incoming requests. As a result, the initial review may take up to 10-15 business days.

[greptile-apps]

Greptile Summary

This PR makes a broad set of improvements to the stock-tracker extension: new UX indicators (pre/post-market icons, ★ for favorited symbols, last-updated timestamp), several bug fixes (formatMoney for zero/negative values, moveDown no-op, move-guard for missing symbols), and robustness improvements (native fetch, getSetCookie() fix, proactive cookie refresh, abort on unmount, corrupt-storage recovery). The architectural refactor lifting all hooks to Command() is clean and correct.

• The CHANGELOG entry uses a hardcoded date (2026-04-30) instead of the required {PR_MERGE_DATE} placeholder — this must be changed before merge.

Confidence Score: 4/5

Safe to merge after fixing the CHANGELOG date placeholder.

One P1 finding: the CHANGELOG entry uses a hardcoded date instead of {PR_MERGE_DATE}, violating the repository's required convention. All source code changes are well-reasoned and an improvement over the previous implementation.

extensions/stock-tracker/CHANGELOG.md — fix the hardcoded date to {PR_MERGE_DATE}.

Important Files Changed

Filename	Overview
extensions/stock-tracker/CHANGELOG.md	New [Improvements] entry uses a hardcoded date instead of the required {PR_MERGE_DATE} placeholder.
extensions/stock-tracker/src/yahoo-finance/client.ts	Switches to native fetch, fixes Set-Cookie parsing with getSetCookie(), threads AbortSignal, adds proactive 12h cache expiry, tightens retry to 401/403; getCookie is missing a response.ok check unlike getCrumb.
extensions/stock-tracker/src/use-stock-info.ts	Adds lastUpdated, stable symbolsKey memo, cleanup abort on unmount, and better error handling with finally; logic is correct.
extensions/stock-tracker/src/favorites-store.ts	Unifies moveUp/moveDown into a single move(delta) helper, guards against missing symbols, and adds safe JSON parsing for corrupted storage.
extensions/stock-tracker/src/index.tsx	Lifts all hooks to the top-level command, eliminating duplicate LocalStorage loads and providing a single source of truth for favorites/search state.
extensions/stock-tracker/src/stock-list-item.tsx	Adds pre/post-market icon accessory and optional ★ favorite indicator; changes are safe.
extensions/stock-tracker/src/yahoo-finance/quote.ts	Expands marketState union to include PREPRE/POSTPOST/CLOSED; error typed as unknown.
extensions/stock-tracker/package.json	Removes cross-fetch, upgrades @raycast/api and dev tools, adds platforms: ["macOS"].

Prompt To Fix All With AI

Fix the following 2 code review issues. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 2
extensions/stock-tracker/CHANGELOG.md:3
**Hardcoded date should be `{PR_MERGE_DATE}`**

The changelog entry uses a hardcoded date (`2026-04-30`) instead of the `{PR_MERGE_DATE}` template variable. The placeholder is automatically replaced with the actual merge date when the PR is merged — using a real date bypasses that mechanism.

```suggestion
## [Improvements] - {PR_MERGE_DATE}
```

### Issue 2 of 2
extensions/stock-tracker/src/yahoo-finance/client.ts:108-121
**`getCookie` does not check `response.ok`**

`getCrumb` was updated to check `response.ok` and throw a descriptive error, but `getCookie` was not given the same treatment. If `https://fc.yahoo.com` returns a non-2xx status, the code silently falls through to parsing `Set-Cookie` headers and then throws the generic "Failed to fetch cookie" message, making it hard to diagnose the real failure.

```suggestion
async function getCookie(signal: AbortSignal): Promise<string> {
  const response = await fetch("https://fc.yahoo.com", { headers: HEADERS, signal });
  if (!response.ok) {
    throw new Error(`Failed to fetch cookie: ${response.status} ${response.statusText}`);
  }
```

_{Reviews (4): Last reviewed commit: "Update CHANGELOG.md and add platforms fi..." | Re-trigger Greptile}

[MeleMannen]

Good changes! LGTM!

[0xdhrv]

Looks good to me, approved ✅

[github-actions]

Published to the Raycast Store:
https://raycast.com/hmarr/stock-tracker

[raycastbot]

🎉 🎉 🎉

Such a great contribution deserves a reward, but unfortunately we couldn't find your Raycast account based on your GitHub username (@vinayh).

Please link your GitHub account to your Raycast account to receive your credits and soon be able to exchange them for some swag.

[greptile-apps]

Hardcoded date should be {PR_MERGE_DATE}

The changelog entry uses a hardcoded date (2026-04-30) instead of the {PR_MERGE_DATE} template variable. The placeholder is automatically replaced with the actual merge date when the PR is merged — using a real date bypasses that mechanism.

Suggested change

	## [Improvements] - 2026-04-30
	## [Improvements] - {PR_MERGE_DATE}

Rule Used: What: Changelog entries must use {PR_MERGE_DATE}... (source)

Prompt To Fix With AI

This is a comment left during a code review.
Path: extensions/stock-tracker/CHANGELOG.md
Line: 3

Comment:
**Hardcoded date should be `{PR_MERGE_DATE}`**

The changelog entry uses a hardcoded date (`2026-04-30`) instead of the `{PR_MERGE_DATE}` template variable. The placeholder is automatically replaced with the actual merge date when the PR is merged — using a real date bypasses that mechanism.

```suggestion
## [Improvements] - {PR_MERGE_DATE}
```

**Rule Used:** What: Changelog entries must use `{PR_MERGE_DATE}`... ([source](https://app.greptile.com/review/custom-context?memory=c2214c11-df56-490a-b1c0-09a385df481a))

How can I resolve this? If you propose a fix, please make it concise.