extract-menu▌
doordash.com/extract-menu-5uzqvc · updated May 21, 2026
MDX-style export adds YAML metadata + attribution linking explainx.ai and this canonical listing URL.
Given a DoorDash restaurant URL or restaurant + city query, extract the full menu — every category, every item, with name, price, description, and popular/featured tags. Read-only — never adds to cart or checks out.
| name | extract-menu |
| title | DoorDash Menu Extraction |
| description | >- Given a DoorDash restaurant URL or restaurant + city query, extract the full menu — every category, every item, with name, price, description, and popular/featured tags. Read-only — never adds to cart or checks out. |
| website | doordash.com |
| category | restaurants |
| tags | - doordash - restaurants - menu - delivery - read-only - cloudflare |
| source | 'browserbase: agent-runtime 2026-05-15' |
| updated | '2026-05-15' |
| recommended_method | hybrid |
| alternative_methods | - method: url-param rationale: >- For chain restaurants, /business/{slug}-{businessId}/menu serves an SSR'd HTML page with menu data embedded as JSON-LD and __NEXT_DATA__, bypassing the Cloudflare managed challenge that gates /store/ URLs. Fastest path; ~100× cheaper than browser. Only available for chain brands and serves the chain's template menu, not store-specific pricing. - method: browser rationale: >- For independent restaurants or when per-store pricing/DashPass deltas are required, the /store/{slug}-{storeId}/ URL is mandatory. It's Cloudflare-protected with a managed challenge, so the session must use --verified --proxies. Also requires an address-gate bypass (?pickup=true or fill the modal) and scroll-to-mount for IntersectionObserver-lazy categories. - method: api rationale: >- consumer-mobile-bff.doordash.com exposes /v1/stores/{id}/menu and /v3/stores/{id}/ — verified live but require JWT auth (returns 401 authorization_invalid cookieless). Not usable for anonymous menu extraction. Don't waste time on this surface without a refresh token. |
| verified | false |
| proxies | true |
DoorDash Menu Extraction
Purpose
Given a DoorDash restaurant URL or a (restaurant name, city) pair, return the full menu — every category, every item, with name, price (string + float), description, popular/featured tags, and category section header. Also returns top-level restaurant metadata (canonical name, address line if visible, star rating, store-level URL). Read-only: never adds anything to a cart, never clicks "Add", never starts a checkout, never types payment details.
When to Use
- Building a menu index across a chain (Chipotle, Sweetgreen, etc.) — hit the chain-level
/business/{slug}-{businessId}/menuURL once per brand. - Capturing per-store pricing where it varies by location (DashPass member pricing, surge-day surcharges, holiday menus) — the store-scoped
/store/...URL is required. - Snapshotting a menu for a price-tracking, allergen-tracking, or dietary-search downstream consumer.
- Comparing menus across locations of the same chain (use the chain
/business/...URL for the canonical template, then sample a few/store/...URLs for delivery-price deltas).
Workflow
DoorDash exposes two parallel URL surfaces for the same restaurant menu, with very different anti-bot postures. Always check which surface fits the request first — the chain /business/ URL is ~100× cheaper and bypasses the Cloudflare challenge entirely, but it only exists for chain brands and serves the brand's template menu rather than store-specific pricing.
/store/{slug}-{storeId}/ → store-specific, Cloudflare-challenged
/business/{slug}-{businessId}/menu → chain-level, SSR'd HTML, no challenge
page-service.doordash.com/en-US/store/... → underlying SSR layer (same HTML body)
Step 1 — Decide the surface
| Scenario | Surface |
|---|---|
Input is a /store/{slug}-{id}/ URL with a specific store id | Browser (/store/...) — Step 4 |
Input is a /business/{slug}-{id}/ URL (chain hub) | Direct fetch (/business/.../menu) — Step 2 |
| Input is a chain restaurant name + city, and per-store pricing is not required | Direct fetch (resolve chain businessId via Step 3, then Step 2) |
| Input is a chain name + city, and per-store pricing is required (DashPass, geo-specific items) | Browser (/store/...) — Step 4 |
| Input is an independent (non-chain) restaurant | Browser (/store/...) — Step 4. Independents rarely have a /business/ hub; verify via Step 3 search first. |
Step 2 — Fast path: chain menu via /business/.../menu
browse cloud fetch "https://www.doordash.com/business/{slug}-{businessId}/menu" --allow-redirects --output menu.html
Returns SSR'd HTML, status 200, no Cloudflare challenge (verified across multiple business URLs on 2026-05-15). No --proxies flag is needed and adding --verified is not supported by browse cloud fetch anyway.
Caveat — 1 MB Fetch API ceiling. Business menu HTML is typically 1.0–1.5 MB. browse cloud fetch errors with 502 The response body exceeded the maximum allowed size of 1MB on most production restaurants. Two workarounds:
- Browserbase session +
Page.getResourceContent— open the URL in abrowse cloud sessions createsession (no Verified/proxy needed for/business/.../menu), then read the response body via CDP. The 1 MB limit isbrowse cloud fetch-only; full sessions stream the whole document. - Run the fetch in a Browserbase Function (
browse cloud functions ...). The function executes inside Browserbase's network, returns whatever JSON you serialize, and is not subject to the Fetch API's 1 MB cap.
Once you have the HTML, extract from one of three embedded sources (in order of preference):
<script type="application/ld+json">Schema.orgRestaurant/Menu— DoorDash emits structured-data JSON-LD for the menu sections and items, includinghasMenuSection[],hasMenuItem[],name,description,offers.price,offers.priceCurrency. This is the cleanest extraction surface.<script id="__NEXT_DATA__" type="application/json">— the Next.js page-data blob containing the full hydration tree. Menu data lives underprops.pageProps.<...>.menu.categories[].items[]. Schema changes occasionally; always parse defensively.- HTML scrape (last resort) —
<h2 data-anchor-id="MenuItem-{itemId}">,<span data-anchor-id="MenuItem-Price">, category headers as<h2>inside<div data-anchor-id="StoreMenuList">. Fragile across redesigns.
Step 3 — Resolve a restaurant name → business or store ID
If the caller passes a name + city instead of a URL:
# Search the public sitemap index for a chain hub
browse cloud fetch "https://www.doordash.com/sitemap-business-doordash-index.xml" --output biz_idx.xml
# Pick the sharded sitemap, then grep for the slug
browse cloud fetch "https://cdn.doordash.com/sitemaps/sitemaps/sitemap-doordash-0-business-menu.xml" --output biz_smm.xml
grep -oE "/business/{slug-pattern}-[0-9]+/menu" biz_smm.xml | head -1
Or use browse cloud search "site:doordash.com/business {restaurant name} menu" — fast, returns canonical URL directly. Verified working in trace 2026-05-15 (returned /business/chipotle-mexican-grill-115/ as top hit for "chipotle").
If no /business/ page exists, the restaurant is an independent — fall through to Step 4 with the /store/ URL discovered via browse cloud search "site:doordash.com/store {name} {city}".
Step 4 — Browser fallback for /store/... (store-specific or independent)
The /store/{slug}-{storeId}/ URL is Cloudflare-protected with a managed challenge (cType: 'managed', cZone: 'www.doordash.com'). Cleared 6 KB interstitial HTML on every bare fetch attempt observed on 2026-05-15 with and without --proxies. Requires a full headless browser with Verified + residential proxies to render.
SID=$(browse cloud sessions create --keep-alive --verified --proxies | jq -r '.id')
browse cloud browse env remote
browse cloud browse --connect "$SID" open "https://www.doordash.com/store/{slug}-{storeId}/"
browse cloud browse --connect "$SID" wait load
browse cloud browse --connect "$SID" wait timeout 4000 # Cloudflare JS challenge + menu hydration
Cloudflare challenge: With --verified --proxies, the managed challenge typically auto-solves in 3–6 s. If it does not clear, wait an additional 5 s and check browse cloud browse get url; a stuck challenge keeps ?__cf_chl_tk=... in the URL.
Address gate: First store visit in a fresh session pops a "Set delivery address" modal that blocks the menu DOM. Two strategies:
- Skip via URL — append
?pickup=trueto load the pickup variant. Pickup pricing usually matches delivery and there is no address gate. - Fill the modal —
browse cloud browse fill "input[placeholder='Address']" "{city}, {state}", wait 2 s for autocomplete dropdown, click the first suggestionmenuitem, clickbutton: Save. The session cookie persists for subsequent stores in the samebbsession.
Lazy-rendered categories: DoorDash uses an IntersectionObserver to render category sections as the user scrolls. After the menu DOM mounts:
# Scroll to bottom in steps to trigger all categories
for i in 1 2 3 4 5 6; do
browse cloud browse --connect "$SID" scroll 640 360 0 1200
browse cloud browse --connect "$SID" wait timeout 500
done
browse cloud browse --connect "$SID" snapshot
Extract from the snapshot: Each menu item appears as region: MenuItem-{itemId} with child text refs for name, price, description. Tag badges (Popular, Featured, #1 Most Liked) appear as sibling image or text refs inside the same region — look for the exact strings, they are not in data- attributes.
Per-store JSON shortcut: The page makes a hydration POST to /graphql/storeMenu (operation storeMenu or storepageFeed) carrying the storeId. Reading the response body via browser-trace CDP capture is the cleanest extraction — but you must capture during the page load, not after, and the GraphQL endpoint requires page-context cookies (no out-of-band call works — verified, 401 authorization_invalid from a cookieless POST to https://consumer-mobile-bff.doordash.com/v3/stores/{id}/).
Step 5 — Release the session
browse cloud sessions update "$SID" --status REQUEST_RELEASE
Site-Specific Gotchas
- READ-ONLY. Never click the "Add to cart" or "+" buttons under each menu item. Never proceed to checkout. Stop at the menu snapshot.
- Cloudflare managed challenge on
/store/...— every/store/URL returns a 6 KB interstitial (<title>Just a moment...</title>,cType: 'managed',cZone: 'www.doordash.com') on cookieless requests.browse cloud fetch --proxiesdoes not clear it; only a JS-executing browser with--verified --proxiesdoes. Verified 2026-05-15 across multiple stores and with/without proxies. /business/{slug}-{businessId}/menuis the SEO-friendly SSR path — fully indexed inhttps://www.doordash.com/sitemap-business_menu-doordash-index.xml(5 sharded sitemaps undercdn.doordash.com/sitemaps/), returns 200 OK without Cloudflare challenge. This is the fastest known way to extract a chain's menu.page-service.doordash.comis the underlying SSR layer —https://page-service.doordash.com/en-US/store/{slug}-{id}/serves the same SSR'd HTML body that the public/business/URL renders. Both paths exceed the 1 MB Fetch API ceiling, so directbrowse cloud fetchis impractical without one of the workarounds in Step 2.- Two ID schemes, do not confuse them.
/business/chipotle-mexican-grill-115/uses business-id 115 (one per chain brand)./store/chipotle-mexican-grill-san-francisco-303528/uses store-id 303528 (one per physical location). They are not interchangeable. - Store URLs sometimes have a double-id form —
/store/chipotle-mexican-grill-washington-270882/471923/. The first id is the address/location group; the second is the actual store. Both forms route to the same store page. - Consumer-mobile-bff API requires JWT auth.
https://consumer-mobile-bff.doordash.com/v3/stores/{id}/and/v1/stores/{id}/menureturn401 {"name":"authorization_invalid","message":"Access Denied"}from cookieless requests. Fingerprintable viaX-Shortened-Url-Path: v1-stores-idheader. Don't waste time on the BFF without a refresh token — the Identity service atidentity.doordash.com/auth/token/refreshrate-limits and responds 403 to bare callers. - Address gate on first store visit. The first
/store/load in a fresh session always prompts for a delivery address. Bypass with?pickup=trueor fill the modal once and reuse the session cookie via--keep-alive. - Categories are IntersectionObserver-lazy. Scrolling is required to mount the full menu DOM — six 1200 px scrolls with a 500 ms wait between covers the longest menus observed. Don't rely on a single
snapshotafterwait load. - Tag badges live in DOM text, not attributes.
Popular,Featured,#1 Most Liked,Customer Favoriteappear as sibling spans/images inside the item region, not asdata-tagattributes. Match the exact strings. - Asterisks / price suffixes. Some items display "$13.65*" or "$13.65+" —
*indicates "starting at" for items with required modifiers,+indicates a base price with optional add-ons. Strip when emittingprice_float, preserve inprice(string), and flag withflags: ["base_price"]. - Sold-out items render with a strikethrough. They have an
aria-disabled="true"attribute on the item region. Emit them as{ available: false }rather than silently dropping — the caller may need the snapshot for a price database. - Regional locale prefixes —
/en-CA/...,/en-AU/...,/en-NZ/...,/en-GB/...and/fr-CA/...exist. The defaultwww.doordash.com/(no locale) serves US. International stores show currency-localized prices; preserve the currency code fromoffers.priceCurrencyin the JSON-LD or__NEXT_DATA__. m.doordash.comreturns 500 — there is no usable mobile-web subdomain. Don't waste time probing it.browse cloud fetch1 MB ceiling — DoorDash store and business HTML routinely exceeds 1 MB. The Fetch API errors with502 The response body exceeded the maximum allowed size of 1MB. Use a real session for any full-page extraction, or run the fetch inside a Browserbase Function where the limit does not apply.- CDP egress restriction on some sandbox tenants. During skill development on 2026-05-15 the runtime sandbox could resolve
api.browserbase.com(REST API for sessions/fetch/search) but notconnect.usw2.browserbase.com(WSS CDP endpoint), which made livebrowse cloud browse --connectand the autobrowse evaluator unreachable from that sandbox. If a future caller hits the same DNS REFUSED onconnect.{region}.browserbase.com, run the browser portion from a host with unrestricted egress; the API-only paths (Steps 2, 3) work fine from a restricted sandbox. - Cloudflare
__cf_bmcookie persistence. Once a Verified session clears the challenge, the__cf_bmcookie (path/, domaindoordash.comandwww.doordash.com, ~30 min expiry) carries it across/store/...navigations. Keep the session alive (--keep-alive) and reuse it for batch extraction across stores in the same brand.
Expected Output
{
"success": true,
"source": "business_menu",
"restaurant": {
"name": "Chipotle Mexican Grill",
"business_id": 115,
"store_id": 303528,
"url": "https://www.doordash.com/store/chipotle-mexican-grill-san-francisco-303528/",
"business_url": "https://www.doordash.com/business/chipotle-mexican-grill-115/menu",
"address": "525 Market St, San Francisco, CA 94105",
"rating": 4.6,
"rating_count": 12048,
"price_tier": "$",
"cuisines": ["Mexican", "Fast Food", "Bowls"]
},
"categories": [
{
"name": "Popular Items",
"items": [
{
"id": "item-901827",
"name": "Burrito Bowl",
"price": "$13.65",
"price_float": 13.65,
"currency": "USD",
"description": "Your choice of freshly grilled meat, sofritas, or guacamole, and up to five toppings.",
"tags": ["Popular"],
"flags": [],
"available": true,
"image_url": "https://img.cdn4dd.com/p/.../burrito-bowl.jpg"
}
]
},
{
"name": "Tacos",
"items": [
{
"id": "item-901831",
"name": "Three Tacos",
"price": "$11.95+",
"price_float": 11.95,
"currency": "USD",
"description": "Three soft or crispy tacos with your choice of fillings.",
"tags": [],
"flags": ["base_price"],
"available": true
}
]
}
],
"extracted_at": "2026-05-15T23:00:00Z",
"error_reasoning": null
}
Failure shapes:
// Cloudflare challenge stuck (didn't clear after Verified + proxy attempt)
{ "success": false, "error_reasoning": "cloudflare_challenge_unsolved", "url": "..." }
// Address gate not bypassable (no autocomplete match for given city)
{ "success": false, "error_reasoning": "address_gate_no_match", "city": "..." }
// Restaurant not on DoorDash
{ "success": false, "error_reasoning": "restaurant_not_found", "query": "..." }
// Store closed / no menu available
{ "success": true, "restaurant": { ... }, "categories": [], "error_reasoning": "store_closed_or_no_menu" }
How to use extract-menu on Cursor
AI-first code editor with Composer
Prerequisites
Before installing skills in Cursor, ensure your development environment meets these requirements:
- ›Cursor installed and configured on your development machine
- ›Node.js version 16.0+ with npm package manager (verify with
node --version) - ›Active project directory or workspace where you want to add extract-menu
Execute installation command
Execute the skills CLI command in your project's root directory to begin installation:
The skills CLI fetches extract-menu from GitHub repository doordash.com/extract-menu-5uzqvc and configures it for Cursor.
Select Cursor when prompted
The CLI will show a list of available agents. Use arrow keys to navigate and space to select Cursor:
Verify installation
Confirm successful installation by checking the skill directory location:
Reload or restart Cursor to activate extract-menu. Access the skill through slash commands (e.g., /extract-menu) or your agent's skill management interface.
Security & Verification Notice
We perform automated surface-level scans (Gen AI Scanner, Socket, Snyk) during installation. These checks detect common vulnerabilities but do not guarantee complete security. Always review skill source code and verify the publisher's reputation before production use.
Skills execute code in your development environment. Always verify the publisher's identity, review recent commits, and test in isolated environments before production deployment.
List & Monetize Your Skill
Submit your Claude Code skill and start earning
Use Cases▌
Task Automation & Efficiency
Automate repetitive workflows and reduce manual effort
Example
Generate reports, summarize documents, draft communications
Save 3-5 hours per week on routine tasks
Knowledge Enhancement
Learn new skills, understand complex topics, get expert guidance
Example
Explain concepts, provide examples, suggest learning resources
Accelerate learning and skill development by 2x
Quality Improvement
Enhance output quality through reviews, suggestions, and refinements
Example
Review drafts, suggest improvements, catch errors
Improve work quality by 30-40% with less effort
Implementation Guide▌
Prerequisites
- ›Claude Desktop or compatible AI client with skill support
- ›Clear understanding of task or problem to solve
- ›Willingness to iterate and refine outputs
Time Estimate
15-45 minutes depending on use case complexity
Installation Steps
- 1.Install skill using provided installation command
- 2.Test with simple use case relevant to your work
- 3.Evaluate output quality and relevance
- 4.Iterate on prompts to improve results
- 5.Integrate into regular workflow if valuable
Common Pitfalls
- ⚠Expecting perfect results without iteration
- ⚠Not providing enough context in prompts
- ⚠Using skill for tasks outside its intended scope
- ⚠Accepting outputs without review and validation
Best Practices▌
✓ Do
- +Start with clear, specific prompts
- +Provide relevant context and constraints
- +Review and refine all outputs before using
- +Iterate to improve output quality
- +Document successful prompt patterns
✗ Don't
- −Don't use without understanding skill limitations
- −Don't skip validation of outputs
- −Don't share sensitive information in prompts
- −Don't expect skill to replace human judgment
💡 Pro Tips
- ★Be specific about desired format and style
- ★Ask for multiple options to choose from
- ★Request explanations to understand reasoning
- ★Combine AI efficiency with human expertise
When to Use This▌
✓ Use When
Use when skill capabilities match your task, clear ROI on time saved, and you can validate outputs. Best for repetitive tasks, learning, and quality improvement.
✗ Avoid When
Avoid when task requires deep expertise you can't validate, involves sensitive decisions, or when learning process is more valuable than speed of completion.
Learning Path▌
- 1Familiarize yourself with skill capabilities and limitations
- 2Start with low-risk, non-critical tasks
- 3Progress to more complex and valuable use cases
- 4Build expertise through regular use and experimentation
Discussion
Product Hunt–style comments (not star reviews)- No comments yet — start the thread.
Ratings
4.6★★★★★31 reviews- ★★★★★Daniel Taylor· Dec 12, 2024
extract-menu reduced setup friction for our internal harness; good balance of opinion and flexibility.
- ★★★★★Carlos Thompson· Nov 3, 2024
I recommend extract-menu for anyone iterating fast on agent tooling; clear intent and a small, reviewable surface area.
- ★★★★★Alexander Bansal· Oct 22, 2024
Useful defaults in extract-menu — fewer surprises than typical one-off scripts, and it plays nicely with `npx skills` flows.
- ★★★★★Li Haddad· Sep 13, 2024
extract-menu has been reliable in day-to-day use. Documentation quality is above average for community skills.
- ★★★★★Pratham Ware· Sep 1, 2024
Registry listing for extract-menu matched our evaluation — installs cleanly and behaves as described in the markdown.
- ★★★★★Alexander Bhatia· Sep 1, 2024
extract-menu reduced setup friction for our internal harness; good balance of opinion and flexibility.
- ★★★★★Oshnikdeep· Aug 20, 2024
extract-menu reduced setup friction for our internal harness; good balance of opinion and flexibility.
- ★★★★★Carlos Garcia· Aug 20, 2024
Registry listing for extract-menu matched our evaluation — installs cleanly and behaves as described in the markdown.
- ★★★★★Chen Kim· Aug 4, 2024
Solid pick for teams standardizing on skills: extract-menu is focused, and the summary matches what you get after install.
- ★★★★★Benjamin Johnson· Jul 27, 2024
extract-menu reduced setup friction for our internal harness; good balance of opinion and flexibility.
showing 1-10 of 31