# Rule 18: Custodian Staff Parsing from LinkedIn Company Pages

**When manually registering heritage custodian staff from LinkedIn company "People" pages, use the `parse_custodian_staff.py` script to convert raw text files into structured JSON.**

## Overview

LinkedIn company pages have a "People" section that lists all staff members with their names, job titles, connection degree, and mutual connections. This data is valuable for understanding the heritage sector workforce and building network analysis.

## File Locations

| Type | Location |
|------|----------|
| **Raw input files** | `data/custodian/person/manual_hc/{slug}-{timestamp}.md` |
| **Parsed output files** | `data/custodian/person/{slug}_staff_{timestamp}.json` |
| **Parser script** | `scripts/parse_custodian_staff.py` |

## Input File Format

Raw files are created by copy-pasting from LinkedIn company "People" pages:

```
Collectie Overijssel logo
Collectie Overijssel

Museums, Historical Sites, and Zoos
Zwolle, Overijssel
2K followers
51-200 employees

58 associated members

    Annelien Vos-Keen
    Annelien Vos-Keen
    2nd degree connection · 2nd
    Data Analist / KPI- en procesexpert
    Thomas van Maaren, Bob Coret, and 4 other mutual connections

Martine de Boer
Martine de Boer
2nd degree connection · 2nd
Collectiespecialist bij Collectie Overijssel
...
```

## Usage

```bash
python scripts/parse_custodian_staff.py <input_file> <output_file> \
    --custodian-name "Custodian Name" \
    --custodian-slug "custodian-slug"
```

**Example**:
```bash
python scripts/parse_custodian_staff.py \
    data/custodian/person/manual_hc/collectie_overijssel-20251210T0055.md \
    data/custodian/person/collectie_overijssel_staff_20251210T0055.json \
    --custodian-name "Collectie Overijssel" \
    --custodian-slug "collectie-overijssel"
```

**Dry-run mode** (parse without writing):
```bash
python scripts/parse_custodian_staff.py input.md output.json \
    --custodian-name "Name" --custodian-slug "slug" --dry-run
```

## Output Structure

```json
{
  "custodian_metadata": {
    "custodian_name": "Collectie Overijssel",
    "custodian_slug": "collectie-overijssel",
    "name": "Collectie Overijssel",
    "industry": "Museums, Historical Sites, and Zoos",
    "location": { "city": "Zwolle", "region": "Overijssel" },
    "follower_count": "2K",
    "employee_count": "51-200",
    "associated_members": 58
  },
  "source_metadata": {
    "source_type": "linkedin_company_people_page",
    "registered_timestamp": "2025-12-10T00:55:00Z",
    "registration_method": "manual_linkedin_browse",
    "staff_extracted": 39
  },
  "staff": [
    {
      "staff_id": "collectie-overijssel_staff_0000_annelien_vos_keen",
      "name": "Annelien Vos-Keen",
      "name_type": "full",
      "degree": "2nd",
      "headline": "Data Analist / KPI- en procesexpert",
      "mutual_connections": "Thomas van Maaren, Bob Coret, and 4 other mutual connections",
      "heritage_relevant": true,
      "heritage_type": "D"
    }
  ],
  "staff_analysis": {
    "total_staff_extracted": 39,
    "heritage_relevant_count": 25,
    "heritage_relevant_percentage": 64.1,
    "staff_by_heritage_type": { "A": 4, "D": 1, "E": 1, "M": 18, "S": 1 },
    "staff_by_degree": { "1st": 2, "2nd": 37 },
    "staff_by_name_type": { "abbreviated": 1, "full": 38 },
    "common_roles": { "Medewerker": 7, "Coördinator": 5, "Beheerder": 4 }
  },
  "provenance": {
    "data_source": "LINKEDIN_MANUAL_REGISTER",
    "data_tier": "TIER_3_CROWD_SOURCED"
  }
}
```

## Staff ID Format

```
{custodian_slug}_staff_{index:04d}_{name_slug}
```

**Examples**:
- `collectie-overijssel_staff_0000_annelien_vos_keen`
- `nationaal-archief_staff_0042_afelonne_doek`

## Heritage Type Detection

Staff headlines are analyzed for heritage relevance using GLAMORCUBESFIXPHDNT keywords:

| Code | Type | Keywords |
|------|------|----------|
| A | Archive | archief, archivist, archivaris, nationaal archief |
| M | Museum | museum, curator, conservator, collectie |
| L | Library | library, bibliotheek, librarian |
| D | Digital | digital, data, developer, digitalisering |
| E | Education | university, professor, docent, educatie |
| R | Research | research, onderzoek, historicus |
| ... | ... | ... |

## Name Types

| Type | Description | Example |
|------|-------------|---------|
| `full` | Complete first + last name | "Vincent Robijn" |
| `abbreviated` | Contains single-letter initial | "Fairoesh N.", "A.J. Gevers" |
| `anonymous` | Privacy-hidden profile | "LinkedIn Member" |

## Existing Files

| Custodian | Staff Count | Output File |
|-----------|-------------|-------------|
| Collectie Overijssel | 39 | `collectie_overijssel_staff_20251210T0055.json` |
| Nationaal Archief | 373 | `nationaal_archief_staff_20251209T2354.json` |

## Integration with Other Scripts

This script complements `parse_linkedin_connections.py` (Rule 15):

| Script | Purpose | Input |
|--------|---------|-------|
| `parse_linkedin_connections.py` | Parse PERSON's connections | Individual profile connections |
| `parse_custodian_staff.py` | Parse ORGANIZATION's staff | Company "People" page |

## See Also

- Rule 15: Connection Data Registration
- Rule 14: Exa MCP LinkedIn Profile Extraction
- `AGENTS.md` - Complete agent instructions