Luma Events Scraper - Public Calendars & Event Changes

SOURCE COVERAGE & PRACTICAL LIMITS

Full product guide

Documentation captured from build 1.0.4. Example outputs retain their original sample context; they are not live results.

Keep public event calendars in view

For event discovery and calendar monitoring: turn public Luma city and organizer calendar subscriptions into event rows with stable IDs, dates and locations. This reads public calendar exports, without guest lists or private attendee data.

Open this Actor in Console · First run · Results · Cost · Full guide

Try this Actor

• First run: Use San Francisco Public Luma Events below, with 10 saved results and a 100-event calendar window. Inspect the first output before enabling saved comparisons.
• Read the results: Open the default Dataset for events and SUMMARY for calendar coverage. The dates, source event link and changed fields let you inspect a saved schedule change.
• Keep in mind: Exports can include past events and are bounded by the selected window. Keyword matching is literal; a missing event does not prove cancellation or deletion.

San Francisco Public Luma Events — copy this input into the JSON editor in Console.

{
  "calendarUrls": [
    "https://luma.com/sf"
  ],
  "maxResults": 10,
  "maxEventsPerCalendar": 100
}

Result preview

The dataset view highlights these columns:

• Event ID · eventId
• Event title · title
• Calendar · calendarName
• Start date · startDate
• Starts at · startsAt
• Ends at · endsAt

Use the original source links and the field definitions below when interpreting the results. CSV works well for flat columns; JSON preserves nested values.

Usage and pricing

Base result rate on the FREE tier: $3 per 1,000 event observations. The Pricing tab shows your plan's result rate and any Actor Start charge. Set a small result limit for the first run; the requested limit is not a guarantee of that many results.

Full guide

Turn Luma's public city and organizer calendar subscriptions into event records with stable IDs, normalized times, public locations and source links. Add literal topic filters or compare saved observations to detect changes in schedules, titles, status, location and exported descriptions.

Useful for a city event watchlist, an organizer calendar export, or a recurring event research workflow. No API key, login, proxy or browser is required.

Quick start

Run with {} to save up to 20 events visible in the San Francisco public calendar feed. The Console prefill uses the same source. Events are ordered by the exported start date ascending; organizer feeds can include historical events. Use startDate when you only want dates on or after a specific day.

{
  "calendarUrls": ["https://luma.com/sf"],
  "keywords": ["AI", "founder"],
  "maxResults": 20,
  "maxEventsPerCalendar": 100
}

Each saved row is one unique exported event. Pricing is $0.003 per saved event ($3 per 1,000). Duplicates, filtered events, unchanged rows skipped by onlyChanges, and error diagnostics are not saved as paid results. Spending limits can stop a run before its requested window is finished.

Supported sources and coverage

• Public city pages such as https://luma.com/sf and https://luma.com/nyc.
• Public organizer calendar pages such as https://luma.com/ignitegtm.
• Public subscriptions in the form https://api.lu.ma/ics/get?entity=calendar&id=cal-... or entity=discover&id=discplace-....

The Actor resolves the public page's calendar identifier and reads the supported iCalendar export described in Luma's subscription help. A general one-segment event page can be syntactically accepted as an input URL, but will fail with a clear unsupported-page error when it is not a calendar/city. Private calendars, personal entity=user feeds, tokens and event detail paths are unsupported. Source robots rules and public network restrictions apply to every request and redirect.

This is the currently visible subscription window, not all events or complete history. During source verification, a city page reported 70 events while its subscription exported 50. The source may retain past events or omit events. Disappearance is never declared a cancellation or deletion. SUMMARY.sourceCoverage reports exported, selected, rejected and saved counts, source refresh hints and any page-reported count.

The export publishes a 12-hour refresh interval. Schedule no more frequently than twice daily unless the publisher changes its interval. A repeated fetch does not promise fresher underlying content.

Inputs

Output and date semantics

Recurring rules requiring recurrence expansion are skipped and reported as partial coverage; explicit recurrence instances can be identified. Invalid dates, broken intervals and malformed records also appear in SUMMARY, never as event rows. No ticket prices, attendee lists, guest counts or private contact details are promised or extracted.

Repeated runs and reliable history

Enable comparePrevious to receive new, changed or unchanged; otherwise status is not_requested. onlyChanges saves new/changed observations only. The first comparison run warns that it is establishing observations. A first row marked new describes the comparison result; SUMMARY.historyUpdated confirms whether the history write succeeded.

History uses the named store maydit-luma-public-calendar-history, scoped by snapshotName, sorted source URLs, keywords and date filters. Only confirmed saved rows advance it. Other observations remain intact on partial runs, limits and filtered windows. No output row means no new baseline for that identity. Volatile DTSTAMP, SEQUENCE and observation time do not trigger changes.

Use one concurrent writer per history scope. A reread detects some concurrent changes, but it is not a transactional lock. Malformed/incompatible history fails clearly; a new snapshotName starts fresh. History is bounded at 20,000 identities per scope.

Run status and limitations

SUMMARY in the default key-value store records failures, rejected entries, window limits, unchanged/duplicate counts and stops caused by time or spending limits. A requested result/window cap is a successful bounded run; it is not complete historical coverage. Source failures produce partial when useful observations survived. If every source fails and nothing usable is saved, the run fails after writing its summary. Empty legitimate feeds or unmatched filters can succeed with a warning and zero rows.

Requests have a 25-second limit and bounded retries for transient failures. There is no login or challenge bypass. Exported source text and locations can be incomplete or stale; verify event details at the source before attending or making decisions. Public export access does not grant blanket republication rights.

Ready-to-run examples

The following exact inputs also appear in examples.json.

San Francisco Public Luma Events

Export 10 events visible in the San Francisco public Luma calendar feed, with source links and normalized times.

Use the copyable first-run input above.

New York Public Luma Calendar Events

Read 10 visible New York events from Luma's documented public calendar subscription, without private guest or ticket data.

{
  "calendarUrls": [
    "https://luma.com/nyc"
  ],
  "maxResults": 10,
  "maxEventsPerCalendar": 100
}

Ignite Community Luma Calendar - Change Watch

Compare 10 visible Ignite Community calendar events with previously saved observations. The first run establishes a baseline; the export can contain historical events.

{
  "calendarUrls": [
    "https://luma.com/ignitegtm"
  ],
  "maxResults": 10,
  "maxEventsPerCalendar": 100,
  "comparePrevious": true,
  "snapshotName": "ignite-community-demo"
}

Help with your run

Open the Issues tab on this Actor with the run URL, the input field that caused trouble and what you expected to receive. Remove API keys and private information before posting.

Explore more Mayd-It Actors

Understand your run

Open Run report in the run’s Output tab for saved-record counts, reported source issues and limits, and suggested next steps. Use CSV export to take the saved results into a spreadsheet. The existing JSON results and detailed machine-readable summary remain available.

The report reflects this run’s recorded coverage. A successful run or a saved result does not establish that every possible source record was collected.