Australian History MCP Server

Last updated: December 2025

Search Australian archives the easy way. Ask your AI to find newspapers from 1803, aerial photos from 1928, species records, heritage buildings, or government datasets - no manual searching required.

What You Can Ask Your AI to Search For

Instead of navigating 11 different archive websites, just ask your AI:

• "Find newspaper articles about the 1939 bushfires in Victoria"
• "Show me historical aerial photos of Melbourne from the 1950s"
• "What heritage buildings are in Carlton?"
• "Get species sightings of platypus in Tasmania since 2020"
• "Find Prime Ministerial speeches mentioning 'economic reform'" (uses FTS5 full-text search)
• "Search for gold rush artefacts in the National Museum"
• "Find historical placenames within 50km of Ballarat" (uses point+radius)
• "What films are related to Mad Max?" (uses related records)
• "Save this search so I can run it again later" (uses saved queries)
• "Plan a research strategy for the Melbourne Olympics 1956" (uses research planning)
• "Start a session to track my research on early Melbourne" (uses session management)
• "Compress these results to save context" (uses context compression)

Your AI handles the API calls, pagination, and formatting - you just ask questions in plain English.

New in v1.0.0: Research Workflow Tools

• Research Planning - Analyse topics, generate search strategies, identify historical name variations
• Session Management - Track queries, avoid duplicates, resume after context resets
• Context Compression - Reduce accumulated results by 70-85% to stay within token limits
• Checkpoints - Save/restore research progress for long investigations

Quick Start

Run directly with npx - no installation required:

npx @littlebearapps/australian-history-mcp

That's it! 10 of 11 data sources work immediately with no API key. Only Trove requires a free API key.

→ Next: Add to your MCP client via Configuration

Table of Contents

• What You Can Ask Your AI to Search For
• Quick Start
• Data Sources - What You Can Find
• Trove API Key - Why & How to Apply
• Important Notice - Third-Party Data Sources
• Add this MCP to Your AI Agent
• How It Works
• Data Sources - Tools & Examples
• Frequently Asked Questions
• Rate Limits
• Licensing Notes
• Resources
• Contributing
• License

Data Sources - What You Can Find

[!IMPORTANT] Most sources work immediately with no registration. Only Trove requires a free API key.

| Source | Content | |--------|---------| | 🏛️ Public Record Office Victoria (PROV) | Victorian state archives: photos, maps, council records, court files, immigration | | 📰 *Trove (National Library of Australia)** | 1,500+ partner collections: newspapers 1803-1954+, gazettes, books, images, magazines, diaries. See all partner collections → | | 📍 Gazetteer of Historical Australian Placenames (GHAP) | 330,000+ historical placenames with coordinates from ANPS and community datasets via TLCMap | | 🦘 Museums Victoria | Museum objects, natural specimens, species info, educational articles | | 🌿 Atlas of Living Australia (ALA) | 165M+ records including historical museum specimens (1800s-1900s), contemporary citizen science, 153,000+ species profiles | | 🏛️ National Museum of Australia (NMA) | 85,000+ museum objects, people, organisations, places, media | | 🏚️ Victorian Heritage Database (VHD) | 12,000+ heritage places, 700+ shipwrecks, architectural styles | | 🎬 Australian Centre for the Moving Image (ACMI) | 42,000+ films, TV, videogames, digital art, creator info | | 🎤 PM Transcripts | 26,000+ Prime Ministerial speeches, media releases, interviews | | 🖼️ IIIF | Generic manifest/image tools for any IIIF-compliant institution | | ✈️ Geoscience Australia Historical Aerial Photography (GA HAP) | 1.2M+ historical aerial photos (1928-1996), all states/territories |

Want another data source added? Share your idea in Discussions

Trove API Key - Why & How to Apply

Trove is Australia's largest digital research portal, aggregating content from 1,500+ partner organisations including state libraries, the National Archives, Australian War Memorial, universities, museums, and research institutions. A free API key unlocks access to all of this.

Extra Data Sources Available via Trove API Key

With just one API key, you gain access to collections from:

| Category | Notable Partners | What You Can Find | |----------|-----------------|-------------------| | 📚 State Libraries | State Library Victoria, State Library NSW, State Library Queensland, and all other state/territory libraries | Historical photographs, manuscripts, maps, local history, family history records | | 🏛️ National Archives | National Archives of Australia | Immigration records, military service files, government photographs, policy files | | ⚔️ War Memorial | Australian War Memorial | WWI & WWII photographs, unit diaries, service records, military art | | 🎓 Universities | ANU, Melbourne, Sydney, Monash, UNSW, and 30+ other universities | Research papers, theses, academic journals, institutional archives | | 🖼️ Museums | National Gallery, Powerhouse, Australian Museum, state museums | Art collections, object documentation, scientific specimens | | 🔬 Research | CSIRO, AIATSIS, NFSA | Scientific publications, Indigenous collections, film & sound archives | | ⚖️ Government | High Court, Federal Court, Parliament | Legal judgments, parliamentary papers, government publications | | 📰 Newspapers | NLA Digitised (1803-1954+) | 35 million+ newspaper articles with full-text search and OCR |

One key, 1,500+ collections - Instead of registering with dozens of institutions individually, Trove aggregates them all. Use the nuc parameter to filter by specific partners (e.g., nuc: "AWM" for War Memorial, nuc: "VSL" for State Library Victoria).

How to Apply

[!IMPORTANT] Before applying, review the Trove API Terms of Use to understand the requirements and ensure your intended use is eligible.

1. Apply at: https://trove.nla.gov.au/about/create-something/using-api
2. Select "Level 1" (personal/research use)
3. Approval typically within 1 week
4. Add TROVE_API_KEY to your MCP configuration (see Configuration below)

Important Notice - Third-Party Data Sources

[!CAUTION] This MCP server provides programmatic access to third-party public APIs that we do not own or control.

Please be aware:

• Terms may change: Each data source has its own terms of use and API policies that may change at any time without notice.
• Access may change: API endpoints, rate limits, data availability, or access requirements could be modified or discontinued by the source providers.
• User responsibility: You are responsible for reviewing and complying with each data source's terms of use, licensing requirements, and acceptable use policies.
• No warranty: We make no guarantees about the availability, accuracy, or completeness of data from these sources.

Review the terms of use for each data source you access:

| Source | Terms of Use | |--------|--------------| | PROV | Copyright Statement | | Trove | API Terms of Use | | GHAP/TLCMap | Terms of Use | | Museums Victoria | Legals | | ALA | Terms of Use | | NMA | Conditions of Use | | VHD | Disclaimer | | ACMI | Terms & Conditions | | PM Transcripts | Disclaimer | | GA (HAP) | Copyright | | IIIF | Varies by institution - check manifest attribution |

Add this MCP to Your AI Agent

Copy the configuration for your preferred AI client:

Claude Desktop

[!NOTE] Works on all plans including Free. This MCP runs locally via npx, so no paid subscription is required.

Add to your claude_desktop_config.json:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "australian-history": {
      "command": "npx",
      "args": ["-y", "@littlebearapps/australian-history-mcp"],
      "env": {
        "TROVE_API_KEY": "your-trove-api-key"
      }
    }
  }
}

Restart Claude Desktop after saving the file.

Claude Code

Option 1 - CLI command (recommended):

claude mcp add australian-history -- npx -y @littlebearapps/australian-history-mcp

To include the Trove API key:

claude mcp add australian-history -e TROVE_API_KEY=your-trove-api-key -- npx -y @littlebearapps/australian-history-mcp

Option 2 - Config file:

Add to .mcp.json in your project directory:

{
  "mcpServers": {
    "australian-history": {
      "command": "npx",
      "args": ["-y", "@littlebearapps/australian-history-mcp"],
      "env": {
        "TROVE_API_KEY": "your-trove-api-key"
      }
    }
  }
}

Cursor

Add to Cursor Settings → MCP → Add Server, or edit ~/.cursor/mcp.json:

{
  "mcpServers": {
    "australian-history": {
      "command": "npx",
      "args": ["-y", "@littlebearapps/australian-history-mcp"],
      "env": {
        "TROVE_API_KEY": "your-trove-api-key"
      }
    }
  }
}

Codex CLI

Add to your ~/.codex/config.toml:

[mcp_servers.australian-history]
command = "npx"
args = ["-y", "@littlebearapps/australian-history-mcp"]
env = { "TROVE_API_KEY" = "your-trove-api-key" }

Gemini CLI

Add to your ~/.gemini/settings.json:

{
  "mcpServers": {
    "australian-history": {
      "command": "npx",
      "args": ["-y", "@littlebearapps/australian-history-mcp"],
      "env": {
        "TROVE_API_KEY": "your-trove-api-key"
      }
    }
  }
}

VS Code

Add to .vscode/mcp.json or user MCP configuration:

{
  "servers": {
    "australian-history": {
      "command": "npx",
      "args": ["-y", "@littlebearapps/australian-history-mcp"],
      "env": {
        "TROVE_API_KEY": "your-trove-api-key"
      }
    }
  }
}

How It Works

This server uses dynamic tool loading - instead of exposing all 75 data tools upfront, it presents just 22 meta-tools organised by function:

Core Tools (10)

| Meta-Tool | Purpose | |-----------|---------| | tools | Discover available data tools by keyword, source, or category | | schema | Get full parameters for a specific tool | | run | Execute any data tool by name | | search | Search across multiple sources at once | | open | Open URLs in your browser | | export | Export results to CSV, JSON, or Markdown | | save_query | Save a named query for later reuse | | list_queries | List saved queries with filtering options | | run_query | Execute a saved query with optional overrides | | delete_query | Remove a saved query by name |

Research Planning Tools (1)

| Meta-Tool | Purpose | |-----------|---------| | plan_search | Analyse topic, generate search strategy, create plan.md |

Session Management Tools (7)

| Meta-Tool | Purpose | |-----------|---------| | session_start | Start a named research session | | session_status | Get current progress and coverage gaps | | session_end | End session with final report | | session_resume | Resume a paused or previous session | | session_list | List all sessions with optional filters | | session_export | Export session data (JSON, Markdown, CSV) | | session_note | Add notes to current session |

Context Compression Tools (4)

| Meta-Tool | Purpose | |-----------|---------| | compress | Reduce records to essential fields (70-85% token savings) | | urls | Extract only URLs from records | | dedupe | Remove duplicate records using URL and title matching | | checkpoint | Save/load/list/delete research checkpoints |

Why? This reduces token usage by 86% (~1,600 vs ~11,909 tokens), making your AI more efficient. Your AI discovers what tools are available, loads parameters only when needed, and executes searches on your behalf.

Token Efficiency for Long Research Sessions

When researching a topic across multiple searches, results accumulate and consume context. The compression tools help manage this:

| Compression Level | Tokens per Record | Use Case | |-------------------|-------------------|----------| | minimal | ~20 | Just IDs and URLs for bookmarking | | standard | ~50 | Title, year, source for review | | full | ~80 | All metadata except descriptions |

Example workflow:

# Plan your research
plan_search(topic="Melbourne Olympics 1956")

# Start tracking session
session_start(name="olympics-research", topic="...")

# Searches are automatically logged
search(query="Melbourne Olympics", sources=["trove", "prov"])

# Compress results to save context (70-85% reduction)
compress(records=results, level="standard")

# End session with summary
session_end()

💡 For backwards compatibility: Set MCP_MODE=legacy to expose all 75 tools directly.

Data Sources - Tools & Examples

Tools:

| Tool | Description | |------|-------------| | prov_search | Search Victorian state archives with category filter (Agency, Series, Item, etc.) | | prov_get_images | Extract image URLs from a PROV digitised record via IIIF manifest | | prov_harvest | Bulk download PROV records with pagination | | prov_get_agency | Get agency details by VA number | | prov_get_series | Get series details by VPRS number | | prov_get_items | Get items within a series by VPRS number |

Examples:

# Research: Find digitised historical railway photographs
prov_search with query: "railway", digitisedOnly: true, limit: 50

# Research: Search council meeting minutes from early 1900s
prov_search with query: "council meeting", dateFrom: "1900", dateTo: "1920"

# Technical: Extract all images from a digitised record
prov_get_images with manifestUrl: "<manifest-url-from-search>", size: "full"

⚠️ API Key Required: Trove tools require a free API key. Apply for your key before using these tools.

Tools:

| Tool | Description | |------|-------------| | trove_search | Search Trove with sortby, filters, holdings for newspapers, gazettes, images, books | | trove_newspaper_article | Get full article details including OCR text and PDF link | | trove_list_titles | List available newspaper or gazette titles by state | | trove_title_details | Get title details with available years and issue counts | | trove_harvest | Bulk download Trove records with cursor-based pagination | | trove_get_contributor | Get contributor details by NUC code | | trove_list_contributors | List/search all 1500+ contributing libraries | | trove_list_magazine_titles | List available magazine titles | | trove_get_magazine_title | Get magazine title details with years/issues | | trove_get_work | Get book/image/map/music details by ID (with holdings, links, versions) | | trove_get_versions | Get all versions of a work with holdings information | | trove_get_person | Get person/organisation biographical data | | trove_get_list | Get user-curated research list by ID | | trove_search_people | Search people and organisations |

Examples:

# Research: Find 1930s newspaper articles about Melbourne floods
trove_search with query: "Melbourne flood", category: "newspaper",
  dateFrom: "1930", dateTo: "1939", state: "vic"

# Research: Search State Library Victoria collections only
trove_search with query: "gold rush", category: "image", nuc: "VSL"

# Technical: Sort results by date (oldest first) with full text
trove_search with query: "bushrangers", sortby: "dateasc",
  includeFullText: true, limit: 100

Partner Collections Available via Trove:

Trove aggregates content from 1,500+ partner organisations across Australia. Use the nuc parameter to filter by contributing institution:

| Partner Type | Example NUC Codes | Content | |--------------|-------------------|---------| | State Libraries | VSL, SLNSW, QSL, SLSA, SLWA, TLIB | Photographs, manuscripts, maps, local history | | National Archives | NAA | Government records, immigration files, photographs | | War Memorial | AWM | Military history, photographs, unit records | | Universities | ANU, UMEL, UNSW, USYD, UQ + :IR suffix for repositories | Research papers, theses, academic collections | | Museums | NMA, NGA, NGV, MAAS, AM, MV | Objects, art collections, documentation | | Research | CSIRO, AIATSIS, NFSA | Scientific publications, Indigenous collections, film/sound |

# Search State Library Victoria photographs
trove_search with query: "Melbourne", category: "image", nuc: "VSL"

# Search War Memorial collections
trove_search with query: "Gallipoli", nuc: "AWM"

# Browse all contributing libraries
trove_list_contributors with query: "university"

# See which partners have content for your search
trove_search with query: "gold rush", includeFacets: true,
  facetFields: ["partnerNuc"]

Note: NUC filtering works for image, book, magazine, research, diary, music categories. Newspaper/gazette content is NLA-digitised without per-article NUC data. See docs/quickrefs/trove-partners.md for the complete partner guide.

Tools:

| Tool | Description | |------|-------------| | ghap_search | Search historical placenames by name, state, LGA, bounding box, or point+radius | | ghap_get_place | Get place details by TLCMap ID | | ghap_list_layers | List all available community data layers | | ghap_get_layer | Get all places from a specific data layer | | ghap_harvest | Bulk download placename records with filters |

Examples:

# Research: Find historical places named "Melbourne"
ghap_search with query: "Melbourne", state: "VIC"

# Research: Search for placenames in a specific Local Government Area
ghap_search with query: "creek", lga: "Yarra"

# Technical: Get all places from a community-contributed layer
ghap_get_layer with layerId: 123

Tools:

| Tool | Description | |------|-------------| | museumsvic_search | Search Museums Victoria with random sort option for discovery | | museumsvic_get_article | Get an educational article by ID | | museumsvic_get_item | Get a museum object (photograph, artefact, technology) by ID | | museumsvic_get_species | Get species info (taxonomy, biology, habitat, distribution) by ID | | museumsvic_get_specimen | Get a natural science specimen with taxonomy and collection info | | museumsvic_harvest | Bulk download Museums Victoria records with pagination |

Examples:

# Research: Find platypus species information
museumsvic_search with query: "platypus", recordType: "species"

# Research: Search gold rush era museum objects
museumsvic_search with query: "gold rush", category: "history & technology"

# Technical: Discover random specimens with images
museumsvic_search with recordType: "specimen", hasImages: true, random: true

Tools:

| Tool | Description | |------|-------------| | ala_search_occurrences | Search species occurrence records by taxon, location, date, or point+radius | | ala_search_species | Search species by common or scientific name | | ala_get_species | Get species profile with taxonomy, images, distribution | | ala_harvest | Bulk download occurrence records with pagination | | ala_search_images | Search images by keyword, taxon, or species | | ala_match_name | Resolve taxonomic names to classification | | ala_list_species_lists | List user-curated species lists | | ala_get_species_list | Get species list details by druid |

Examples:

# Research: Find koala sightings in Victoria since 2020
ala_search_occurrences with scientificName: "Phascolarctos cinereus",
  stateProvince: "Victoria", startYear: 2020

# Research: Get detailed species profile with images
ala_get_species with guid: "<species-guid-from-search>"

# Technical: Search images of eucalyptus species
ala_search_images with query: "Eucalyptus", limit: 50

Tools:

| Tool | Description | |------|-------------| | nma_search_objects | Search museum collection objects by keyword, type, date | | nma_get_object | Get detailed object record with media | | nma_search_places | Search places of significance in the collection | | nma_harvest | Bulk download collection records with pagination | | nma_get_place | Get place details by ID | | nma_search_parties | Search people and organisations | | nma_get_party | Get party (person/org) details by ID | | nma_search_media | Search images, video, and sound | | nma_get_media | Get media details by ID | | nma_get_related | Get related objects, parties, and places for a record |

Examples:

# Research: Find boomerang artefacts in the collection
nma_search_objects with query: "boomerang", limit: 20

# Research: Search for historical photographs
nma_search_objects with query: "convict", type: "Photographs"

# Technical: Find people and organisations related to exploration
nma_search_parties with query: "explorer", limit: 30

Tools:

| Tool | Description | |------|-------------| | vhd_search_places | Search Victorian heritage places by name, location, style | | vhd_get_place | Get detailed heritage place info with images | | vhd_search_shipwrecks | Search Victorian shipwrecks along the coast | | vhd_harvest | Bulk download heritage records with pagination | | vhd_get_shipwreck | Get shipwreck details by ID | | vhd_list_municipalities | List all Victorian municipalities | | vhd_list_architectural_styles | List architectural style classifications | | vhd_list_themes | List heritage themes (history, economics, etc.) | | vhd_list_periods | List historical periods |

Examples:

# Research: Find Victorian-era heritage buildings in Melbourne CBD
vhd_search_places with query: "bank", municipality: "MELBOURNE CITY",
  architecturalStyle: "Victorian Period (1851-1901)"

# Research: Search shipwrecks along the Victorian coast
vhd_search_shipwrecks with query: "barque", limit: 20

# Technical: Get detailed heritage place record with history
vhd_get_place with id: 12345

Tools:

| Tool | Description | |------|-------------| | acmi_search_works | Search ACMI collection with field and size options | | acmi_get_work | Get detailed work information by ID | | acmi_harvest | Bulk download ACMI collection works with pagination | | acmi_list_creators | List creators (directors, actors, studios) with pagination | | acmi_list_constellations | List curated thematic collections | | acmi_get_constellation | Get constellation details with works | | acmi_get_related | Get related works (parts, recommendations, group members) |

Examples:

# Research: Find Australian feature films from the 1970s
acmi_search_works with query: "Australian", type: "Film", year: 1975

# Research: Search for classic videogames in the collection
acmi_search_works with query: "arcade", type: "Videogame"

# Technical: Get related works for a film (parts, recommendations)
acmi_get_related with workId: 12345

Tools:

| Tool | Description | |------|-------------| | pm_transcripts_get_transcript | Get Prime Ministerial transcript by ID | | pm_transcripts_harvest | Bulk download transcripts with PM name filter | | pm_transcripts_search | Full-text search across all indexed transcripts (requires local index) | | pm_transcripts_build_index | Build/rebuild the local SQLite FTS5 search index | | pm_transcripts_index_stats | Get index statistics (record count, size, last updated) |

💡 Full-Text Search: The pm_transcripts_search tool uses a local SQLite FTS5 index for fast, powerful searches. Run pm_transcripts_build_index once to create the index (~43 minutes for all 26,000 transcripts). Supports Boolean operators, phrase matching, and BM25 ranking.

⚠️ Harvest Limitation: The PM Transcripts API has no search endpoint, so harvesting scans IDs sequentially. Filtering by PM name can be slow. For targeted PM research, use startFrom near the PM's era:

• Curtin/Chifley (1940s): ~1-2000
• Menzies (1950s-60s): ~2000-4000
• Whitlam/Fraser (1970s-80s): ~4000-5000
• Hawke/Keating (1983-1996): ~5000-10000
• Howard (1996-2007): ~10000-18000
• Rudd/Gillard/Abbott+ (2007+): ~18000-26000

Examples:

# Get a specific transcript
pm_transcripts_get_transcript with id: 12345

# Harvest Hawke era transcripts
pm_transcripts_harvest with primeMinister: "Hawke", maxRecords: 100

# Build the full-text search index (one-time setup)
pm_transcripts_build_index with mode: "full"

# Full-text search for "economic reform" across all transcripts
pm_transcripts_search with query: "economic reform", limit: 20

# Search with phrase matching and PM filter
pm_transcripts_search with query: '"unemployment rate"', primeMinister: "Keating"

Tools:

| Tool | Description | |------|-------------| | iiif_get_manifest | Fetch and parse IIIF manifest from any institution | | iiif_get_image_url | Construct IIIF Image API URLs for various sizes/formats |

Examples:

# Research: Access State Library Victoria digitised content
iiif_get_manifest with manifestUrl:
  "https://rosetta.slv.vic.gov.au/delivery/iiif/presentation/2.1/IE145082/manifest"

# Technical: Construct thumbnail URL for a specific image
iiif_get_image_url with imageServiceUrl: "<url-from-manifest>",
  size: "!200,200", format: "jpg"

# Technical: Get full resolution image URL
iiif_get_image_url with imageServiceUrl: "<url-from-manifest>",
  size: "max", format: "jpg", quality: "default"

Tools:

| Tool | Description | |------|-------------| | ga_hap_search | Search historical aerial photos by state, year, location, bbox, or point+radius | | ga_hap_get_photo | Get photo details by OBJECTID or film/run/frame | | ga_hap_harvest | Bulk download photo records with pagination |

⚠️ Lookup Note: The RUN and FRAME fields are strings (e.g., "COAST TIE 2", "C-KEY"), not integers. For reliable lookups, use objectId from search results rather than film/run/frame combination.

Examples:

# Search 1950s Victorian aerial photos
ga_hap_search with state: "VIC", yearFrom: 1950, yearTo: 1960, scannedOnly: true

# Get photo details by ID
ga_hap_get_photo with objectId: 12345

# Harvest photos by Melbourne bounding box
ga_hap_harvest with bbox: "144.9,-37.9,145.1,-37.7", maxRecords: 100

* Section contains usage notes

Frequently Asked Questions

Use trove_search with category: "newspaper". Filter by date with dateFrom and dateTo parameters (format: YYYY or YYYY-MM-DD), and by state with the state parameter (e.g., "vic", "nsw", "qld").

Only Trove requires an API key. All 10 other sources (PROV, GHAP, Museums Victoria, ALA, NMA, VHD, ACMI, PM Transcripts, IIIF, GA HAP) work immediately with no registration required.

For PROV records, use prov_get_images with the manifest URL from search results. For any IIIF-compliant institution, use iiif_get_manifest to get canvas details, then iiif_get_image_url to construct download URLs in your preferred size and format.

Each source has a _harvest tool for bulk downloads with pagination:

• prov_harvest, trove_harvest, ghap_harvest
• museumsvic_harvest, ala_harvest, nma_harvest
• vhd_harvest, acmi_harvest, pm_transcripts_harvest, ga_hap_harvest

Yes. Three sources support point+radius spatial queries using lat, lon, and radiusKm parameters:

• ala_search_occurrences - Species sightings within radius of a point
• ga_hap_search - Aerial photos within radius of a point
• ghap_search - Historical placenames within radius of a point

For bounding box queries, use bbox parameter (format: "minLon,minLat,maxLon,maxLat") with ga_hap_search or ghap_search.

For state/region filtering: ALA supports stateProvince (e.g., "Victoria"), VHD supports municipality for Victorian LGAs, and GA HAP supports state codes (VIC, NSW, etc.).

Use ala_search_species for scientific or common names, or museumsvic_search with recordType: "species". Get detailed profiles with ala_get_species using the GUID from search results. ALA covers all Australian species; Museums Victoria focuses on Victorian fauna and flora.

Dynamic mode (default) exposes 22 meta-tools organised into 4 categories:

• Core tools (10): tools, schema, run, search, open, export, save_query, list_queries, run_query, delete_query
• Research planning (1): plan_search
• Session management (7): session_start, session_status, session_end, session_resume, session_list, session_export, session_note
• Context compression (4): compress, urls, dedupe, checkpoint

This reduces token usage by 86% (~1,600 vs ~11,909 tokens). Your AI discovers and executes tools on demand.

Legacy mode exposes all 75 data tools directly. Use this if you need backwards compatibility or prefer direct tool access.

Switch modes by setting MCP_MODE=legacy in your configuration environment variables.

Rate Limits

• PROV: No documented rate limit
• Trove: 200 API calls per minute
• GHAP/TLCMap: No documented rate limit
• Museums Victoria: No documented rate limit
• ALA: No documented rate limit
• NMA: No documented rate limit
• VHD: No documented rate limit
• ACMI: No documented rate limit
• PM Transcripts: No documented rate limit (be respectful, 100ms delays recommended)
• GA HAP: No documented rate limit (standard ArcGIS Feature Service)

Licensing Notes

[!NOTE] The licences below apply to content from each source. For API usage terms and conditions, see Important Notice - Third-Party Data Sources.

• PROV: CC-BY-NC (non-commercial use)
• Trove: Terms vary by content contributor; check individual items
• GHAP/TLCMap: CC-BY 4.0 (data contributed to TLCMap)
• Museums Victoria: CC-BY 4.0 (most records), Public Domain (some)
• ALA: Various (data contributors specify); mostly CC-BY
• NMA: CC-BY-NC (non-commercial use)
• VHD: CC-BY 4.0 (Victorian government open data)
• ACMI: CC0 (public domain dedication for API data)
• PM Transcripts: Australian Government (Crown copyright)
• IIIF: Varies by institution (check manifest attribution field)
• GA HAP: CC-BY 4.0 (Geoscience Australia, attribution required)
• This MCP Server: MIT License

Resources

• PROV Collection API
• Trove API v3 Guide
• GHAP/TLCMap Developer Docs
• Museums Victoria Collections API
• ALA Web Services
• National Museum of Australia API
• Victorian Heritage Database
• ACMI Collection API
• PM Transcripts
• IIIF Documentation
• Geoscience Australia HAP
• GLAM Workbench - PROV
• GLAM Workbench - Trove

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT - See LICENSE for details.