Smart Event Scraper — Category, Date Range & Keyword Filter Across 6 Event Platforms

One search, six event platforms, and filters that actually filter. Give it a city, a category, an optional keyword and an optional date window — get back a single de-duplicated list of events from AllEvents, Eventbrite, Meetup, District, EventsEye and ConferenceAlerts, with the events that don't match already discarded.

🎯 This is the filtering flagship of the Techforce Global event family. Single-platform Actors give you everything a site lists and leave the filtering to you. This one applies the category, keyword and date filters during the run, across all six platforms at once.

Browse all Actors by Techforce Global at scraper.techforce.global.

---

Smart Event Scraper vs. doing this yourself

This table exists because there is a real, specific gap — not a marketing one.

	Six separate scrapers / manual search	Smart Event Scraper
Platforms per run	One site at a time, six runs to stitch together	✅ All six in a single run, one dataset
Category filter	⚠️ Each site names its categories differently — you map them by hand	✅ One category input, translated per platform
Keyword search	⚠️ Some sites have search, some don't — you filter the rest yourself	✅ Handled either way, transparently
Date range	❌ Most event sites offer no usable date-range filter	✅ date_from / date_to applied to every source
Duplicates	⚠️ The same conference appears on three platforms	✅ De-duplicated across sources before saving
Wrong-city results	⚠️ Ask a site for "London" and some quietly return their default city	✅ Locations are resolved before scraping; unresolvable ones are skipped, not guessed
A slow platform	❌ One hanging site blocks everything	✅ Each source has its own time budget; a slow one contributes partial results instead of stalling the run
Delivery	Manual export → import	✅ Optional direct delivery to Notion, Slack, Linear and more over MCP

---

AI Discovery & MCP Metadata

ai_discovery:
  actor: "techforce.global/smart-event-scraper"
  slug: "smart-event-scraper"
  category: "events / lead-generation / market-research"
  one_line: >
    Searches six event platforms at once for a given city and returns only the events
    matching the chosen category, keyword and date range - de-duplicated across sources,
    with an optional direct delivery into Notion, Slack, Linear or Google Sheets over MCP.

  primary_capability: "multi-platform event discovery with server-side and client-side filtering"

  data_sources:
    - "AllEvents"
    - "Eventbrite"
    - "Meetup"
    - "District (India only)"
    - "EventsEye (trade shows and exhibitions)"
    - "ConferenceAlerts (academic and professional conferences)"

  input_entities:
    - "location (city name, required)"
    - "category (any | technology | business | food | health | arts | education, required)"
    - "search_query (optional keyword)"
    - "date_from / date_to (optional YYYY-MM-DD window)"
    - "sources (subset of the six platforms)"
    - "limit (per source, 0 = no limit)"
    - "country_code (optional ISO-2 override)"
    - "eventType (Meetup only: inPerson | online)"

  output_entities:
    - "source"
    - "title"
    - "url"
    - "date"
    - "city"
    - "venue"
    - "price"
    - "category"
    - "search_query (only when a keyword was used)"
    - "filter_date_from / filter_date_to (only when a date filter was used)"

  supported_tasks:
    - "find-events-by-category-in-a-city"
    - "find-events-matching-a-keyword"
    - "find-events-within-a-date-range"
    - "compare-event-coverage-across-platforms"
    - "build-an-event-calendar-for-a-city"
    - "find-trade-shows-and-exhibitions"
    - "find-academic-conferences"
    - "deliver-matched-events-into-notion-slack-linear-sheets"

  unsupported_tasks:
    - "attendee-lists-or-attendee-contact-details"
    - "ticket-purchasing-or-booking"
    - "organiser-email-addresses-or-phone-numbers"
    - "multiple-cities-in-one-run"
    - "past-events-or-historical-archives"
    - "event-descriptions-or-agendas-or-speaker-lists"
    - "images-banners-or-attachments"

  limit_is_per_source: >
    `limit` caps each scraping task, NOT the run. With a specific category selected the
    Actor runs one task per (source x category), so the run total can be several times
    `limit` before de-duplication. Set `limit` low first, then scale up.

  category_any_is_the_fast_path: >
    category "any" runs each source once with no category filter and keeps every event.
    It is both the broadest and the fastest option. A specific category runs more tasks.

  location_is_single_city: >
    One city per run. There is no multi-city input - schedule or fan out one run per city.

  district_is_india_only: >
    District covers Indian cities only. For any other location it is skipped with a warning
    rather than returning events from the wrong city.

  keyword_behaviour: >
    Platforms with their own search engine receive the keyword server-side and their results
    are trusted as-is (a semantic engine legitimately returns titles without the literal word).
    Platforms without search are filtered on the keyword locally.

  known_limitations:
    price_is_often_unavailable: "many platforms do not publish price on listing pages - expect 'N/A'"
    date_format_varies: "`date` is returned as the source published it; formats differ per platform"
    no_event_descriptions: "listing-level fields only - no description, agenda or speaker list"
    single_city_per_run: "one `location` per run"
    unresolvable_city_skips_meetup: "if the city's country cannot be resolved, Meetup is skipped unless country_code is set"

  when_not_to_use: >
    You need attendee data, organiser contact details, ticket purchasing, event descriptions,
    several cities in one run, or historical/past events.

  mcp_delivery:
    modes: ["summary", "perEvent", "chunked", "none"]
    default: "summary"
    event_placeholders: ["{title}", "{url}", "{date}", "{city}", "{venue}", "{price}", "{category}", "{source}", "{event}"]
    run_placeholders: ["{location}", "{categories}", "{sources}", "{searchQuery}", "{dateFrom}", "{dateTo}", "{eventCount}", "{sourceBreakdown}", "{events}", "{part}", "{partCount}", "{message}"]
    never_blocks_the_dataset: >
      Delivery runs after every event is already saved. A bad tool name, an unauthorised
      connector or an upstream outage is logged and skipped - it never costs you the data.

---

{
    "@context": "https://schema.org",
    "@type": "SoftwareApplication",
    "name": "Smart Event Scraper",
    "alternateName": "Smart Event Scraper — Category, Date Range & Keyword Filter Across 6 Event Platforms",
    "applicationCategory": "BusinessApplication",
    "applicationSubCategory": "Web Scraping / Event Data Aggregation",
    "operatingSystem": "Any (cloud-hosted on Apify)",
    "description": "Searches AllEvents, Eventbrite, Meetup, District, EventsEye and ConferenceAlerts in a single run and returns only the events matching the chosen city, category, keyword and date range — de-duplicated across platforms, with an optional direct delivery into Notion, Slack, Linear or Google Sheets over MCP.",
    "url": "https://apify.com/techforce.global/smart-event-scraper",
    "offers": {
        "@type": "Offer",
        "category": "Pay per event",
        "description": "Billed per event pushed to the dataset. Current rates are shown on the Actor's Apify page.",
        "url": "https://apify.com/techforce.global/smart-event-scraper"
    }
}

---

What this Actor does

• Searches six event platforms in one run. AllEvents, Eventbrite, Meetup, District, EventsEye and ConferenceAlerts — general listings, trade shows and academic conferences in the same dataset. Pick a subset when you only care about some of them.

• Filters while it scrapes, not after. The category, keyword and date window are applied during the run. What lands in your dataset is what matched — you don't post-process a dump.

• Resolves the location before it scrapes. Several event sites accept a city they don't actually cover and quietly return their default city's events instead. This Actor resolves the city first (roughly 190 cities across 55+ countries) and skips a source it can't resolve rather than handing you events from the wrong place.

• De-duplicates across platforms. The same conference listed on Eventbrite, Meetup and AllEvents becomes one row, not three.

• Never lets one slow platform hold up the run. Each source has its own time budget and results are saved as each source finishes. A platform that stalls contributes whatever it had collected; the rest of the run is unaffected.

• Optionally delivers straight into your tools. Set an MCP connector and the matched events are pushed into Notion, Slack, Linear, Google Sheets or any other connector authorised on your Apify account — no export step.

---

When to use it / when not to use it

Use this Actor when you need…	Supported?
Tech / business / food / health / arts / education events in a city	✅ Yes — category
Every event type, no category filter	✅ Yes — category: "any" (also the fastest)
Events matching a keyword ("AI workshop", "React")	✅ Yes — search_query
Events inside a specific date window	✅ Yes — date_from / date_to
Trade shows and exhibitions	✅ Yes — the eventseye source
Academic and professional conferences	✅ Yes — the conferencealerts source
Online rather than in-person Meetup events	✅ Yes — eventType: "online"
A cross-platform coverage comparison for one city	✅ Yes — the By source dataset view
Events pushed straight into Notion / Slack / Linear	✅ Yes — MCP delivery
Several cities in one run	❌ One location per run — run once per city
Attendee lists or attendee contact details	❌ Never collected
Organiser emails or phone numbers	❌ Not returned
Full event descriptions, agendas or speaker lists	❌ Listing-level fields only
Ticket purchasing or seat availability	❌ Not a booking tool
Past events / historical archives	❌ Upcoming listings only
Event images or banners	❌ Not returned

---

Execution Rules — Avoid Common Agent Failures

1. limit is per source, not per run. It caps each scraping task. With a specific category selected the Actor runs one task per (source × category), so the run total can be several times limit before de-duplication. Start at limit: 10, check the count, then scale.

2. category is required and "any" is a real value. "any" means no category filter: every source runs once, every event is kept, and each row is labelled "any". It is both the broadest option and the fastest. An empty or invalid category makes the run exit immediately with a "Please select at least one valid category" status.

3. location is one city. No lists, no regions, no countries. Use the plain city name (London, New York, Ahmedabad) — not London, UK.

4. Leave country_code empty. It is detected from the city. Only set it if the run log tells you a city's country could not be resolved. Setting it wrong overrides a correct detection and returns events from another country (the log warns when this happens).

5. District only covers Indian cities. For any other location it is skipped with a warning. That is deliberate — it would otherwise return events from a different Indian city.

6. Dates must be YYYY-MM-DD. The filter compares against the date the source published, and formats vary per platform; an unparseable event date means that event is not excluded by the filter. Treat the date window as a strong filter, not a guarantee.

7. With a specific category, EventsEye and District are keyword-classified. They only publish city-wide listings, so their events are matched against your selected category and dropped if none matches. If you want everything they list, use category: "any".

8. A keyword is not a substring match on every source. Platforms with their own search engine handle it server-side and their results are trusted — a search for networking can legitimately return "Women in Tech Summit". Platforms without search are filtered locally on the literal terms.

9. Free Apify plans are capped at 1 result per source. This is a plan limit, not a bug. The run finishes successfully and says so in its status message.

10. MCP delivery needs both mcpConnector and mcpTool. A connector without a tool name logs a warning and skips delivery. If the tool name is wrong, the log lists every tool the connector actually offers — read it and retry.

11. This Actor drives real browsers. Give the run at least 2 GB of memory. Below that, browser-backed sources are liable to be killed mid-run.

---

Input Contract

Parameter	Type	Required	Default	Description
location	string	yes	"New York"	Single city name. London, New York, Ahmedabad.
category	string	yes	"any"	any, technology, business, food, health, arts, education. any = no category filter (fastest).
search_query	string	no	""	Keyword, e.g. AI workshop, React meetup. Empty = no keyword filter.
sources	array<string>	no	all six	Any subset of allevents, eventbrite, meetup, district, eventseye, conferencealerts.
date_from	string	no	""	YYYY-MM-DD. Keep events on or after this date.
date_to	string	no	""	YYYY-MM-DD. Keep events on or before this date.
limit	integer	no	10	Max events per scraping task (0–100). 0 = no limit.
country_code	string	no	""	ISO-2 override (gb, us, in). Leave empty — auto-detected.
eventType	string	no	"inPerson"	inPerson or online. Meetup only; ignored elsewhere.

Delivery (optional)

Parameter	Type	Default	Description
mcpConnector	string	""	Apify MCP connector ID — notion, slack, linear, google-sheets, … Empty = dataset only.
mcpTool	string	""	Tool to call on that connector. Required when a connector is set.
deliveryMode	string	"summary"	summary (one call per run), perEvent, chunked, none.
mcpArguments	object	{}	Arguments for the tool. {placeholder} tokens are substituted per call.
mcpMessageTemplate	string	""	Text rendered into {message}. Same placeholders.

Delivery placeholders

Scope	Placeholders
Per event (perEvent mode)	{title} {url} {date} {city} {venue} {price} {category} {source} {event}
Per run (all modes)	{location} {categories} {sources} {searchQuery} {dateFrom} {dateTo} {eventCount} {sourceBreakdown} {events} {message}
Chunked mode only	{part} {partCount}

{event} is a pre-formatted block for one event; {events} is the same for the whole set (or the current chunk). {sourceBreakdown} renders as allevents: 12, meetup: 7. An unknown placeholder renders as empty text rather than failing the call. Fields the source published as N/A render as empty, so templates don't fill up with the literal string.

Valid sources values

Value	Platform	Notes
allevents	AllEvents	General listings, has native search
eventbrite	Eventbrite	General listings, has native search. Price detail improves with an optional Eventbrite API token
meetup	Meetup	Community events. Honours eventType. Skipped if the city's country can't be resolved
district	District	Indian cities only — skipped elsewhere
eventseye	EventsEye	Trade shows and exhibitions. No native search — keyword filtered locally
conferencealerts	ConferenceAlerts	Academic and professional conferences, has native search

---

Output Contract

One dataset item per matched event.

Field	Type	Nullable	Meaning
source	string	no	Which platform produced it: allevents, eventbrite, meetup, district, eventseye, conferencealerts.
title	string	no	Event name as published. "N/A" if the platform exposed none.
url	string	no	Direct link to the event page. "N/A" if unavailable.
date	string	no	Date as the source published it — ISO timestamp, plain date or a range, depending on platform. "N/A" if absent.
city	string	no	City the event belongs to.
venue	string	no	Venue or place name. "N/A" for online and unlisted venues.
price	string	no	Price as published. "N/A" is common — most listing pages don't show one.
category	string	no	Category matched under, or "any" when no category filter was applied.
search_query	string	yes	Present only when a keyword was supplied.
filter_date_from	string	yes	Present only when date_from was supplied.
filter_date_to	string	yes	Present only when date_to was supplied.

Every non-nullable field is always present. Missing data is the string "N/A", never null and never an absent key — so row["venue"] is always safe.

Example output object

{
  "source": "meetup",
  "title": "AI Builders London — Agents in Production",
  "url": "https://www.meetup.com/ai-builders-london/events/301884492/",
  "date": "2026-09-17T18:30:00+00:00",
  "city": "London",
  "venue": "Runway East, Borough",
  "price": "Free",
  "category": "technology",
  "search_query": "AI",
  "filter_date_from": "2026-09-01",
  "filter_date_to": "2026-09-30"
}

What a successful run looks like

• Status: SUCCEEDED
• Dataset: one item per matched event, saved incrementally as each source finishes
• Log: a per-source progress line, then a final count with how many events were excluded by the date filter
• Zero results is also a success. A narrow keyword, a tight date window, a small city or a subset of sources can all legitimately match nothing. The run still succeeds with an empty dataset — check the log to see which sources ran and which were skipped.

Dataset views

View	Shows
Events	Every matched event with date, venue, price, category, source and link
By source	The same events grouped for coverage comparison across platforms
Filters applied	Which keyword and date window each event was matched against

---

Invocation & Integration

Apify Console

Open the Actor page, fill in the Input tab, click Start.

MCP — add as an agent tool

$claude mcp add apify --transport http https://mcp.apify.com/?actors=techforce.global/smart-event-scraper

Then ask your agent for "technology events in London in September" and it will call this Actor with the right input.

Python

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")

run = client.actor("techforce.global/smart-event-scraper").call(run_input={
    "location": "London",
    "category": "technology",
    "search_query": "AI",
    "date_from": "2026-09-01",
    "date_to": "2026-09-30",
    "sources": ["allevents", "eventbrite", "meetup", "conferencealerts"],
    "limit": 25,
})

for event in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(event["date"], "|", event["title"], "|", event["source"])

JavaScript / TypeScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });

const run = await client.actor('techforce.global/smart-event-scraper').call({
    location: 'London',
    category: 'technology',
    search_query: 'AI',
    date_from: '2026-09-01',
    date_to: '2026-09-30',
    limit: 25,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.table(items.map(({ title, date, source }) => ({ title, date, source })));

cURL — synchronous (waits, returns the dataset)

curl -X POST "https://api.apify.com/v2/acts/techforce.global~smart-event-scraper/run-sync-get-dataset-items?token=<YOUR_APIFY_TOKEN>" \
  -H 'Content-Type: application/json' \
  -d '{
        "location": "London",
        "category": "technology",
        "date_from": "2026-09-01",
        "date_to": "2026-09-30",
        "limit": 25
      }'

cURL — asynchronous with a spend cap

curl -X POST "https://api.apify.com/v2/acts/techforce.global~smart-event-scraper/runs?token=<YOUR_APIFY_TOKEN>&maxTotalChargeUsd=2" \
  -H 'Content-Type: application/json' \
  -d '{ "location": "London", "category": "any", "limit": 50 }'

maxTotalChargeUsd is the hard ceiling for the run. Use it on any unattended or scheduled run.

With MCP delivery — matched events straight into Notion

{
  "location": "London",
  "category": "technology",
  "date_from": "2026-09-01",
  "date_to": "2026-09-30",
  "limit": 25,
  "mcpConnector": "notion",
  "mcpTool": "notion-create-page",
  "deliveryMode": "summary",
  "mcpArguments": {
    "parent": { "database_id": "<YOUR_NOTION_DATABASE_ID>" },
    "title": "Tech events in {location} — {dateFrom} to {dateTo}",
    "content": "{eventCount} events found ({sourceBreakdown})\n\n{events}"
  }
}

Switch deliveryMode to perEvent for one Notion page per event, or chunked when the run returns more events than a single page comfortably holds.

---

Autonomous Agent Error Handling & Resolution Matrix

Each row is a condition observable from the run's status message, log or dataset.

Condition	What you see	Cause	Deterministic action
NO_VALID_CATEGORY	Status Skipped: Please select at least one valid category. — empty dataset	category was missing, empty or not a valid value	Modify Input: set category to any or one of the six values. Do not retry unchanged.
DISTRICT_SKIPPED	Log warns District only covers Indian cities	location is not an Indian city District serves	Expected. Drop district from sources for non-Indian cities to remove the warning.
MEETUP_SKIPPED_UNKNOWN_COUNTRY	Log warns the country for the city could not be resolved; Meetup returns nothing	City is outside the recognised city list	Modify Input: set country_code to the correct ISO-2 code and retry.
COUNTRY_CODE_OVERRIDE_WARNING	Log warns your country_code overrides the known country for that city	country_code disagrees with the detected country	Modify Input: clear country_code unless you meant it. Results may be from another country.
ZERO_RESULTS	SUCCEEDED, empty dataset	Filters too narrow, small city, or the chosen sources had nothing	Widen, then retry once: drop search_query, widen the date window, or use category: "any".
SOURCE_TIMED_OUT	Log notes a source timed out and kept partial results	That platform was slow or unresponsive this run	Retry later, or exclude the source. Other sources are unaffected — data already saved is valid.
FEWER_THAN_LIMIT	Dataset smaller than limit × sources	De-duplication, category classification and date filtering all remove rows	Expected. limit is a ceiling per task, not a quota.
FREE_PLAN_CAPPED	Status says up to 1 result per source; tiny dataset	Free Apify plan	Upgrade the plan. Retrying will not change it.
EVENTBRITE_PRICE_UNAVAILABLE	Log notes no Eventbrite API token; price is "N/A"	Optional price enrichment not configured	Optional: set the EVENTBRITE_API_TOKEN environment variable for richer price data.
CONNECTOR_TOOL_NOT_FOUND	Log: tool not available on this connector, followed by the available tools	mcpTool name is wrong	Modify Input: copy an exact tool name from that log line. Dataset is unaffected.
CONNECTOR_NO_TOOL	Log: a connector is selected but no tool name was provided	mcpConnector set, mcpTool empty	Modify Input: set mcpTool, or clear mcpConnector.
CONNECTOR_UNREACHABLE	Log: could not reach the connector	Connector not authorised on the account, or running outside the Apify platform	Authorise the connector in Apify, or set deliveryMode: "none". Data is already saved.
MEMORY_KILLED	Run fails or a source dies mid-way	Run memory too low for browser-backed sources	Raise memory to at least 2 GB and retry.

Verify any platform-level HTTP status behaviour against the Apify API reference before relying on it in automation — the rows above describe this Actor's observable behaviour, not Apify's transport layer.

---

Natural Language → Actor Mapping

A user says…	Intent	Input
"Tech events in London next month"	Category + city + window	{"location":"London","category":"technology","date_from":"2026-09-01","date_to":"2026-09-30"}
"What's happening in Chicago this week?"	Everything, narrow window	{"location":"Chicago","category":"any","date_from":"2026-08-25","date_to":"2026-08-31"}
"Find AI workshops in New York"	Keyword	{"location":"New York","category":"technology","search_query":"AI workshop"}
"Business networking events in Mumbai"	Category + city	{"location":"Mumbai","category":"business"}
"Trade shows in Frankfurt"	Exhibitions only	{"location":"Frankfurt","category":"any","sources":["eventseye"]}
"Academic conferences in Singapore"	Conferences only	{"location":"Singapore","category":"any","sources":["conferencealerts"]}
"Online React meetups"	Keyword + online	{"location":"London","category":"technology","search_query":"React","sources":["meetup"],"eventType":"online"}
"Food festivals in Austin in October"	Category + window	{"location":"Austin","category":"food","date_from":"2026-10-01","date_to":"2026-10-31"}
"Art exhibitions in Berlin"	Category	{"location":"Berlin","category":"arts"}
"Just Eventbrite and Meetup, Ahmedabad, health"	Source subset	{"location":"Ahmedabad","category":"health","sources":["eventbrite","meetup"]}
"Give me everything in Delhi, no filters"	Broadest run	{"location":"Delhi","category":"any","limit":100}
"Put next month's London tech events in our Notion"	Scrape + deliver	Above, plus mcpConnector, mcpTool, mcpArguments
"Post today's Bangalore startup events to Slack"	Scrape + deliver	{"location":"Bangalore","category":"business","search_query":"startup","mcpConnector":"slack","mcpTool":"slack-post-message","deliveryMode":"summary"}

Should NOT route here: who is attending an event · organiser or attendee email addresses and phone numbers · buying tickets or checking seat availability · full event descriptions, agendas or speaker lists · several cities in one run · past events or historical archives · event images or banners · venue reviews or ratings · a single platform scraped exhaustively with no filters (use that platform's dedicated Actor).

---

Agent Execution & Routing Logic

Does the user want UPCOMING EVENTS for ONE city?
├── NO → not this Actor.
│        attendees/contacts → no Actor in this family returns them
│        tickets/booking    → not a scraping task
│        many cities        → one run per city (schedule or fan out)
│        past events        → not available
│
└── YES
    │
    ├── Do they name a category or interest?
    │     ├── yes → category: technology | business | food | health | arts | education
    │     └── no  → category: "any"        (broadest AND fastest)
    │
    ├── Do they name a keyword or topic?  → search_query: "<keyword>"
    │
    ├── Do they name a time window?       → date_from / date_to as YYYY-MM-DD
    │
    ├── Do they name a kind of event?
    │     ├── trade shows / exhibitions   → sources: ["eventseye"]
    │     ├── academic conferences        → sources: ["conferencealerts"]
    │     ├── community meetups           → sources: ["meetup"]  (+ eventType)
    │     ├── Indian city, local listings → sources include "district"
    │     └── unspecified                 → leave sources at all six
    │
    ├── Is the city outside the common list?  → set country_code (ISO-2)
    │
    ├── Set limit: start at 10, confirm the count, then scale (it is PER TASK)
    │
    └── Should the results land in another app?
          ├── yes → mcpConnector + mcpTool + deliveryMode
          │           one item per event   → perEvent
          │           one digest per run   → summary
          │           too many for one doc → chunked
          └── no  → leave mcpConnector empty; dataset only

---

Cost Control & Pricing Transparency

This Actor bills per event pushed (pay-per-event), not per minute of compute. You pay for the events that actually matched your filters — a run that finds nothing costs nothing in event charges. Current per-event rates are shown on the Actor's Apify page; the Apify Console shows the live total while a run is in progress.

What drives the cost of a run:

Choice	Effect on cost
limit	The main lever. It caps each scraping task, so the run ceiling is roughly limit × number of tasks.
category: "any" vs a specific category	"any" runs one task per source. A specific category runs one task per (source × category) — more tasks, more potential events.
sources	Fewer sources → fewer tasks → fewer events. Pick only the platforms you need.
search_query and date window	Narrower filters discard more events before they are charged.
De-duplication	Cross-platform duplicates are removed before saving, so you are not charged twice for the same event.

Keeping runs predictable:

1. Start with limit: 10 and one or two sources. Read the final log count, then scale.
2. Always set a spend cap on unattended runs — maxTotalChargeUsd on the API call, or the run limit in the Console. This is the only hard ceiling.
3. Prefer narrow filters over a big limit. A tight date window with limit: 50 usually costs less and is more useful than limit: 10 with no filters.
4. Free Apify plans are capped at 1 result per source, which makes free runs effectively a smoke test rather than a data pull.

---

Troubleshooting

#	Symptom you see	Most likely cause	Fix
1	Run ends immediately with a Skipped: Please select at least one valid category. status	category was missing, empty, or not one of the seven accepted values	Set category to any or one of the six named values. It is a required field, and an invalid value exits rather than defaulting.
2	Empty dataset, SUCCEEDED status	Filters too narrow, a small city, or the selected sources had nothing for that window	Widen once: drop search_query, open the date window, or switch to category: "any". Then stop — repeated identical retries will not help.
3	Far fewer rows than limit × sources	limit is a per-task ceiling, not a quota, and de-duplication, category classification and date filtering all remove rows afterwards	Expected. Read the final log count rather than predicting from limit.
4	Many more rows than you expected	A specific category runs one task per (source × category); "any" runs one task per source	Use "any" for the widest and fastest run, or lower limit when naming a category.
5	District returned nothing and the log warns about Indian cities	District covers Indian cities only and is skipped elsewhere by design	Expected. Remove district from sources for non-Indian locations to silence the warning.
6	Meetup returned nothing and the log says the city's country could not be resolved	The city is outside the recognised list, so no country could be derived	Set country_code to the correct ISO-2 code (gb, us, in) and retry.
7	Results are from the wrong country	country_code was set and disagrees with the city's detected country — your value wins	Clear country_code unless you meant it. The log warns when the override happens.
8	location: "London, UK" behaves oddly	location takes a plain city name, not a city-plus-country string, and not a list or region	Pass London. Use country_code if disambiguation is genuinely needed.
9	A keyword search returns events that don't contain the keyword	Platforms with their own search engine handle the query server-side and their relevance ranking is trusted	Expected, and usually desirable. Filter title locally afterwards if you need literal matching.
10	EventsEye or District results look category-filtered oddly	Both publish city-wide listings only, so with a named category their events are keyword-classified and dropped when nothing matches	Use category: "any" to keep everything those two list.
11	Events outside the date window still appear	The filter compares against the date the source published, and an event whose date cannot be parsed is not excluded	Treat the window as a strong filter, not a guarantee. Re-filter on date in your consumer if it must be exact.
12	Eventbrite price is "N/A" on every row	The optional EVENTBRITE_API_TOKEN environment variable is not set, so price enrichment is skipped	Optional. Set the variable for richer Eventbrite pricing; everything else works without it.
13	One source returned nothing while the others worked	Each source has its own time budget — a slow platform is cancelled and the run keeps whatever it had	Retry later, or drop that source. Data already saved is valid.
14	Run fails, or a source dies mid-way	Run memory too low for the browser-backed sources	Give the run at least 2 GB. The Actor declares 2–4 GB for this reason.
15	Exactly 1 result per source	Free Apify plan cap	Upgrade the plan. The run finishes successfully and says so in its status message.
16	Connector configured but nothing delivered	mcpTool is empty — a connector without a tool name logs a warning and skips delivery	Set mcpTool. If the name is wrong, the log lists every tool the connector actually offers.
17	Connector errors partway through a large run	perEvent mode makes one call per event	Switch to chunked or summary.

Diagnostic checklist before opening an issue

1. Read the status message first. An input problem produces a Skipped: … terminal status rather than a failure, and it names the reason.
2. Confirm the run's status. SUCCEEDED with zero rows is a valid outcome here, not a crash.
3. Probe narrow: one source, category: "any", limit: 10, no search_query, no dates. If that returns events, the pipeline works and the issue is your filters.
4. Check memory is at least 2 GB before blaming a source that returned nothing.
5. Scan the log for per-source warnings. District skipping, an unresolved country, a country_code override and a source timeout each look identical from the dataset but need different fixes.

If it survives all five, open an Issues ticket with the run ID, the exact input JSON, and what you expected.

---

FAQ

Setup & scope

Do I need accounts or API keys for the six platforms?

No. You need an Apify account and API token only. One optional extra: setting an EVENTBRITE_API_TOKEN environment variable enriches Eventbrite pricing, but every source works without it.

Which platforms does it search?

Six: AllEvents, Eventbrite, Meetup, District (Indian cities only), EventsEye (trade shows) and ConferenceAlerts (academic conferences). Pick a subset in sources for a faster, more focused run.

Is this an official API for any of those platforms?

No. This is an independent Actor and is not affiliated with, endorsed by, or sponsored by any of the six platforms.

Can I plug it into Claude, Cursor, or a LangChain agent?

Yes — it is a native MCP tool:

claude mcp add --transport http apify "https://mcp.apify.com?tools=techforce.global/smart-event-scraper"

How much memory does it need?

At least 2 GB. Some sources are driven by a real browser, and below 2 GB those are liable to be killed mid-run. The Actor declares 2–4 GB.

Filters & input

Is limit per run or per source?

Per task. With category: "any" that is one task per source. With a named category it is one task per (source × category), so the run total can be several times limit before de-duplication removes overlaps.

What does category: "any" actually do?

It applies no category filter: every source runs once, every event is kept, and each row is labelled "any". It is both the broadest option and the fastest.

Can I search more than one city?

No — location is a single plain city name. Loop the Actor per city from your orchestration layer.

Should I set country_code?

Leave it empty. It is detected from the city. Only set it when the log tells you a city's country could not be resolved — setting it wrongly overrides a correct detection and returns events from another country.

Why is eventType listed as Meetup-only?

Because only Meetup exposes an in-person vs online distinction the Actor can pass through. Every other source ignores the field.

Are duplicates removed across platforms?

Yes. The same event listed on two platforms is de-duplicated before saving, so you are not charged twice for it.

Output & data

What do I get per event?

Eleven fields: source, title, url, date, city, venue, price, category, plus search_query, filter_date_from and filter_date_to echoing the filters that produced the row. Three Console views group them: overview, by source, and filters applied.

Why do the rows carry my own filter values back?

So a dataset assembled from several runs stays self-describing — you can tell which query and date window produced any given row without keeping a separate log.

Are the dates normalised to one format?

No. date is captured as each platform presents it, and formats differ between them. Parse before sorting, and expect some values that no parser will accept.

Do I get attendee lists, organiser emails, or ticket availability?

No. Public listing fields only — no personal data and no authenticated pages.

Why is price sometimes "N/A"?

Because the platform did not publish one on the listing, or, for Eventbrite specifically, because the optional price-enrichment token is not configured.

Cost & runtime

How is this billed?

Pay-per-event: you are charged for events pushed to the dataset, not for compute time. A run that matches nothing costs nothing in event charges. Current rates are on the Actor's Apify page.

What drives the cost of a run?

limit first, then the number of sources, then whether you named a category (which multiplies the task count). Narrow filters discard events before they are charged, so a tight date window with a higher limit is often both cheaper and more useful than a low limit with no filters.

What does the free plan give me?

1 result per source — a smoke test to confirm a city and category return anything, rather than a data pull.

How do I guarantee an agent never overspends?

Pass maxTotalChargeUsd as a query parameter on the run endpoint. It is the only hard ceiling, and it is worth setting on every unattended or scheduled run.

---

Companion machine-readable files

File	What it defines
.actor/input_schema.json	Every input, its type, default, allowed values and validation
.actor/dataset_schema.json	Output field types and the three Console views
.actor/output_schema.json	Where the run's output is exposed
.actor/actor.json	Actor identity, memory bounds and environment variables

If this page and the schema files disagree, the schema files win. They are what the platform validates against at run time; this page is documentation and can lag behind a release. Validate generated input against input_schema.json, not against this README.

---

Related Actors by Techforce Global

Actor	Best for
This Actor	Filtered event discovery across six platforms at once — category, keyword, date range
Meetup Scraper	Meetup only, scraped exhaustively for one city
District Events Scraper	District only, all categories for an Indian city
All Events Scraper	AllEvents only, full city listings
Ticketmaster Scraper	Ticketed concerts, sports and shows
Eventim Scraper	European ticketed events

Use a single-platform Actor when you want everything one site lists. Use this Actor when you want only what matches — across all of them.

Browse the full catalogue at scraper.techforce.global.

---

Compliance & Data Privacy

• Public listings only. Everything returned is published on a public event page with no login. Nothing behind authentication is accessed.

• No personal data is collected. No attendee names, no attendee or organiser email addresses, no phone numbers. Output is event metadata: name, date, city, venue, price, category, link. Under GDPR / UK GDPR and CCPA/CPRA, event metadata of this kind is not personal data.

• Organiser names are out of scope. The dataset carries a venue or place name, not an individual's contact details. If a venue field happens to contain a person's name (a sole-trader studio, for example), you become a controller for that field the moment you store it — apply your own retention policy.

• Marketing to organisers is your responsibility. This Actor gives you no contact route, and finding one elsewhere puts you under CAN-SPAM (US), CASL (Canada), ePrivacy/PECR (UK/EU) and India's DPDP Act 2023. Consent and opt-out are the sender's obligation, not the scraper's.

• Each platform's Terms of Service apply to you. AllEvents, Eventbrite, Meetup, District, EventsEye and ConferenceAlerts each set their own terms on automated access and on reusing their listings — including republishing them commercially. Review the terms of the sources you enable and stay within them.

• Attribution and re-publication. Event listings are the platforms' and organisers' content. Link back rather than mirroring wholesale, and don't present a scraped listing as your own original content.

• Data minimisation. Use sources, category, search_query, the date window and limit to collect what you need and nothing more. It is cheaper, faster and the defensible position.

---

🆘 Support & Custom Pipeline Engineering

Need automated multi-city scheduled feeds, webhook enrichment, or enterprise data warehouse syncs?

• Email: bhavin.shah@techforceglobal.com
• Custom Enterprise Integrations: Book a 15-Minute Technical Consultation
• Maintained by: Techforce Global — Specialists in High-Performance Web Scrapers and Agentic Workflows.

---

---

🏷️ Structured data for search & AI discovery

SoftwareApplication describes the Actor; FAQPage wraps the FAQ above so those answers are eligible for rich results, matching it question-for-question.

{
  "@context": "https://schema.org",
  "@type": "SoftwareApplication",
  "name": "Smart Event Scraper",
  "alternateName": "Smart Event Scraper — Category, Date Range & Keyword Filter Across 6 Event Platforms",
  "applicationCategory": "BusinessApplication",
  "applicationSubCategory": "Web Scraping / Event Data Aggregation",
  "operatingSystem": "Any (cloud-hosted on Apify)",
  "url": "https://apify.com/techforce.global/smart-event-scraper",
  "description": "Searches AllEvents, Eventbrite, Meetup, District, EventsEye and ConferenceAlerts in a single run and returns only the events matching the chosen city, category, keyword and date range — de-duplicated across platforms, with an optional direct delivery into Notion, Slack, Linear or Google Sheets over MCP.",
  "author": {
    "@type": "Organization",
    "name": "Techforce Global",
    "url": "https://scraper.techforce.global/"
  },
  "provider": {
    "@type": "Organization",
    "name": "Apify",
    "url": "https://apify.com"
  },
  "offers": {
    "@type": "Offer",
    "category": "Pay per event",
    "description": "Billed per event pushed to the dataset. Current rates are shown on the Actor's Apify page.",
    "url": "https://apify.com/techforce.global/smart-event-scraper"
  },
  "featureList": [
    "Six event platforms searched in one run",
    "Category filter applied during the run, not after",
    "Keyword search handled server-side or locally per platform",
    "Date-range filter across every source",
    "Cross-platform de-duplication, so an event listed twice is charged once",
    "Location resolved before scraping, unresolvable cities skipped rather than guessed",
    "Trade shows and academic conferences alongside general listings",
    "Per-source time budgets so one slow platform never stalls the run",
    "Browser concurrency capped so parallel sources do not starve each other",
    "Each row echoes the query and date window that produced it, keeping multi-run datasets self-describing",
    "Charged per event saved, so a run that matches nothing costs nothing in event charges",
    "Optional Eventbrite price enrichment via an environment variable",
    "Optional MCP delivery to Notion, Slack, Linear, Google Sheets and more",
    "Chunked delivery mode keeps long event lists under connector size and timeout caps",
    "Three Console dataset views including cross-platform coverage"
  ],
  "isAccessibleForFree": false,
  "softwareRequirements": "Apify account and API token; at least 2 GB of run memory",
  "softwareHelp": {
    "@type": "CreativeWork",
    "url": "https://apify.com/techforce.global/smart-event-scraper"
  },
  "keywords": "event scraper, event data API, category filter, date range filter, keyword search, AllEvents scraper, Eventbrite scraper, Meetup scraper, District events, EventsEye trade shows, ConferenceAlerts, event aggregation, MCP connector, event calendar data"
}

{
    "@context": "https://schema.org",
    "@type": "FAQPage",
    "mainEntity": [
        {
            "@type": "Question",
            "name": "Do I need accounts or API keys for the six platforms?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "No. You need an Apify account and API token only. One optional extra: setting an EVENTBRITE_API_TOKEN environment variable enriches Eventbrite pricing, but every source works without it."
            }
        },
        {
            "@type": "Question",
            "name": "Which platforms does it search?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Six: AllEvents, Eventbrite, Meetup, District (Indian cities only), EventsEye (trade shows) and ConferenceAlerts (academic conferences). Pick a subset in sources for a faster, more focused run."
            }
        },
        {
            "@type": "Question",
            "name": "Is this an official API for any of those platforms?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "No. This is an independent Actor and is not affiliated with, endorsed by, or sponsored by any of the six platforms."
            }
        },
        {
            "@type": "Question",
            "name": "Can I plug it into Claude, Cursor, or a LangChain agent?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Yes - it is a native MCP tool. Add it with: claude mcp add --transport http apify \"https://mcp.apify.com?tools=techforce.global/smart-event-scraper\""
            }
        },
        {
            "@type": "Question",
            "name": "How much memory does it need?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "At least 2 GB. Some sources are driven by a real browser, and below 2 GB those are liable to be killed mid-run. The Actor declares 2-4 GB."
            }
        },
        {
            "@type": "Question",
            "name": "Is limit per run or per source?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Per task. With category any that is one task per source. With a named category it is one task per source and category combination, so the run total can be several times limit before de-duplication removes overlaps."
            }
        },
        {
            "@type": "Question",
            "name": "What does category \"any\" actually do?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "It applies no category filter: every source runs once, every event is kept, and each row is labelled any. It is both the broadest option and the fastest."
            }
        },
        {
            "@type": "Question",
            "name": "Can I search more than one city?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "No - location is a single plain city name. Loop the Actor per city from your orchestration layer."
            }
        },
        {
            "@type": "Question",
            "name": "Should I set country_code?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Leave it empty. It is detected from the city. Only set it when the log tells you a city's country could not be resolved - setting it wrongly overrides a correct detection and returns events from another country."
            }
        },
        {
            "@type": "Question",
            "name": "Why is eventType listed as Meetup-only?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Because only Meetup exposes an in-person versus online distinction the Actor can pass through. Every other source ignores the field."
            }
        },
        {
            "@type": "Question",
            "name": "Are duplicates removed across platforms?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Yes. The same event listed on two platforms is de-duplicated before saving, so you are not charged twice for it."
            }
        },
        {
            "@type": "Question",
            "name": "What do I get per event?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Eleven fields: source, title, url, date, city, venue, price and category, plus search_query, filter_date_from and filter_date_to echoing the filters that produced the row. Three Console views group them: overview, by source, and filters applied."
            }
        },
        {
            "@type": "Question",
            "name": "Why do the rows carry my own filter values back?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "So a dataset assembled from several runs stays self-describing - you can tell which query and date window produced any given row without keeping a separate log."
            }
        },
        {
            "@type": "Question",
            "name": "Are the dates normalised to one format?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "No. date is captured as each platform presents it, and formats differ between them. Parse before sorting, and expect some values that no parser will accept."
            }
        },
        {
            "@type": "Question",
            "name": "Do I get attendee lists, organiser emails, or ticket availability?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "No. Public listing fields only - no personal data and no authenticated pages."
            }
        },
        {
            "@type": "Question",
            "name": "Why is price sometimes \"N/A\"?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Because the platform did not publish one on the listing, or, for Eventbrite specifically, because the optional price-enrichment token is not configured."
            }
        },
        {
            "@type": "Question",
            "name": "How is this billed?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Pay-per-event: you are charged for events pushed to the dataset, not for compute time. A run that matches nothing costs nothing in event charges. Current rates are on the Actor's Apify page."
            }
        },
        {
            "@type": "Question",
            "name": "What drives the cost of a run?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "limit first, then the number of sources, then whether you named a category, which multiplies the task count. Narrow filters discard events before they are charged, so a tight date window with a higher limit is often both cheaper and more useful than a low limit with no filters."
            }
        },
        {
            "@type": "Question",
            "name": "What does the free plan give me?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "1 result per source - a smoke test to confirm a city and category return anything, rather than a data pull."
            }
        },
        {
            "@type": "Question",
            "name": "How do I guarantee an agent never overspends?",
            "acceptedAnswer": {
                "@type": "Answer",
                "text": "Pass maxTotalChargeUsd as a query parameter on the run endpoint. It is the only hard ceiling, and it is worth setting on every unattended or scheduled run."
            }
        }
    ]
}