Skip to content
Back to skills

Luma Event Attendees

ASecurity

Find speakers, hosts, and guest profiles at conferences and events on Luma. Two modes - free direct scrape for hosts, or Apify-powered search for full guest profiles with LinkedIn/Twitter/bio.

  • 1,232 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 4, 2026
ai-agentspythongobashapi

Works with

  • api

Security analysis

A96/100
  • mediumInstalls packages at runtime which could introduce malicious dependencies

Pro scans all 5 files and shows the line behind each finding

Scanned September 4, 2026

npx -y skills add gooseworks-ai/goose-skills --skill luma-event-attendees --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Luma Event Attendees?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Luma Event Attendees
[![Security: A β€” Skills Directory](https://www.skillsdirectory.com/api/skills/gooseworks-ai-luma-event-attendees-goose-skills/badge)](https://www.skillsdirectory.com/skills/gooseworks-ai-luma-event-attendees-goose-skills)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: luma-event-attendees
version: 1.0.0
description: Find speakers, hosts, and guest profiles at conferences and events on Luma. Two modes - free direct scrape for hosts, or Apify-powered search for full guest profiles with LinkedIn/Twitter/bio.
tags: [lead-generation]
---

# luma-event-attendees

Find and extract speakers, hosts, and registered guest profiles from Luma events for outreach prospecting.

## Two Modes

### 1. Direct Scrape (free)

Scrapes Luma event pages directly. Gets event metadata + hosts. Guest profiles only if publicly embedded in the page.

```bash
python3 scripts/scrape_event.py https://lu.ma/abc123
```

### 2. Apify Search (paid, recommended for guest lists)

Uses the `lexis-solutions/lu-ma-scraper` Apify actor to search Luma and return full event data including **featured guest profiles** (name, bio, LinkedIn, Twitter, Instagram, website).

```bash
python3 scripts/scrape_event.py --search "AI San Francisco"
```

**Cost:** $29/month flat subscription on Apify.
**Rent:** https://console.apify.com/actors/r5gMxLV2rOF3J1fxu

## Setup

### 1. Apify API Token (required for --search mode)

1. Create account: https://apify.com/
2. Get API token: https://console.apify.com/account/integrations
3. Rent the Luma scraper: https://console.apify.com/actors/r5gMxLV2rOF3J1fxu ($29/mo, 24h free trial)
4. Set token:

```bash
export APIFY_API_TOKEN="apify_api_YOUR_TOKEN_HERE"
# Or create .env file in skill directory
```

### 2. Install Dependencies

```bash
pip3 install requests
```

## Usage

### Direct Scrape (free, hosts only)

```bash
# Single event
python3 scripts/scrape_event.py https://lu.ma/pwciozw0

# Multiple events
python3 scripts/scrape_event.py https://lu.ma/abc https://lu.ma/def

# Export to CSV
python3 scripts/scrape_event.py https://lu.ma/abc --output hosts.csv
```

### Apify Search (guest profiles)

```bash
# Search for AI events in SF
python3 scripts/scrape_event.py --search "AI San Francisco"

# Just list events (don't extract people)
python3 scripts/scrape_event.py --search "SaaS NYC" --events-only

# Export all guests to CSV
python3 scripts/scrape_event.py --search "AI San Francisco" --output guests.csv

# Export as JSON
python3 scripts/scrape_event.py --search "AI SF" --output guests.json --json
```

### Caching

Results cached for 24 hours by default:

```bash
# Force fresh fetch
python3 scripts/scrape_event.py --search "AI SF" --no-cache

# Custom cache duration
python3 scripts/scrape_event.py --search "AI SF" --cache-hours 12
```

### Options Reference

```
Positional:
  urls                    Event URLs to scrape directly (free)

Search:
  --search, -s            Search Luma via Apify (e.g., 'AI San Francisco')
  --events-only           Only list events, don't extract people

Output:
  --output, -o            Output file path (.csv or .json)
  --json                  Output JSON format (default: CSV)

Cache:
  --no-cache              Skip cache, always fetch fresh
  --cache-hours           Cache max age in hours (default: 24)
```

## Output Format

### CSV Columns

| name | event_role | bio | title | company | linkedin_url | twitter_url | instagram_url | website_url | username | event_name | event_date | event_url |
|------|-----------|-----|-------|---------|-------------|-------------|---------------|-------------|----------|------------|------------|-----------|

### What You Get Per Person

- **name** - Full name
- **event_role** - Host, Guest, or Speaker
- **bio** - Luma profile bio
- **linkedin_url** - LinkedIn profile URL
- **twitter_url** - Twitter/X profile URL
- **instagram_url** - Instagram handle
- **website_url** - Personal website
- **username** - Luma username
- **event_name** - Which event they're associated with
- **event_date** - Event date (ISO format)
- **event_url** - Link to the event

## AI Agent Workflow

This skill is designed to be called by an AI agent as part of a prospecting workflow:

### Step 1: Find Events

> "Search Luma for AI and SaaS events in San Francisco"

```bash
python3 scripts/scrape_event.py --search "AI San Francisco" --events-only
```

### Step 2: Extract Guest Profiles

> "Get all guest profiles from those events"

```bash
python3 scripts/scrape_event.py --search "AI San Francisco" --output guests.csv
```

### Step 3: Qualify Against ICP

Ask the agent to filter the CSV:

> "From these guests, find founders/VPs at B2B SaaS companies, 20-200 employees"

### Step 4: Enrich

For qualified leads:
- Look up LinkedIn profiles for role/company details
- Research their companies
- Check for overlapping signals (hiring? recently funded?)

### Step 5: Generate Outreach

> "Draft connection requests for qualified guests. I'll be at [event]. We sell [product] at [price]. Keep it casual."

## Data Access Realities

| Data | Direct Scrape (free) | Apify Search (paid) |
|------|---------------------|---------------------|
| Event metadata | Yes | Yes |
| Hosts/organizers | Yes | Yes |
| Featured guests (public RSVPs) | Sometimes | Yes |
| Full attendee list | No (requires auth) | Partial (public profiles only) |
| Guest LinkedIn/Twitter | Yes (if in page) | Yes |
| Guest bio | Yes (if in page) | Yes |
| Guest email | No | No |

**Note:** Luma events have a `show_guest_list` setting. When disabled, guest profiles aren't publicly accessible. The Apify scraper can still get `featured_guests` for events that have them.

## Example Prompts

**Quick search:**
> "Find AI events in SF this month and get me guest profiles"

**Targeted:**
> "Search Luma for 'SaaS growth' events. Export all guest profiles to CSV. Then qualify against our ICP: VP+ at B2B SaaS, 50-500 employees."

**Full workflow:**
> "Search Luma for AI and developer events in SF. Get all guest profiles. For each person with a LinkedIn, check if they match our ICP (founders/VPs at B2B SaaS, 20-200 employees, Series A-C). Draft pre-event connection requests for the ones I'll see at [event name]. We sell GTM engineering at $10K/month. Output qualified leads to CSV."

## Troubleshooting

### "APIFY_API_TOKEN not set"

```bash
export APIFY_API_TOKEN="your_token_here"
```

### "Apify actor not rented"

Rent the Luma scraper at: https://console.apify.com/actors/r5gMxLV2rOF3J1fxu

### No guests found for an event

- The event may have `show_guest_list` disabled
- Try the --search mode which can access featured_guests
- Some events simply don't have public guest profiles

### "ModuleNotFoundError: requests"

```bash
pip3 install requests
```

## Metadata

```yaml
metadata:
  clawdbot:
    emoji: "🎀"
    requires:
      env: ["APIFY_API_TOKEN"]
      bins: ["python3"]
      packages: ["requests"]
```

---

Built by Goose - Powered by Apify (lexis-solutions/lu-ma-scraper)

Files in this skill

  • README.md1.1 KB
  • SKILL.md6.6 KB
  • scripts/apify_client.py8.1 KB
  • scripts/scrape_event.py11.5 KB
  • skill.meta.json270 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.