canonical: https://jentic.com/apis/nla.gov.au/trove

# Nla Trove API

Jentic publishes the only available OpenAPI specification for Trove API, keeping it validated and agent-ready. Trove is the National Library of Australia's discovery service across hundreds of millions of digitised newspapers, gazettes, magazines, books, images, music, archived websites, and biographical records. The API exposes a single unified search endpoint plus targeted lookups for individual works, newspaper articles, gazette articles, magazine titles, lists, people and organisations, and contributor institutions. Authentication is an API key sent in the X-API-KEY header; keys are free to register at trove.nla.gov.au.

## For AI agents

Search and retrieve digitised Australian newspapers, gazettes, magazines, books, and biographical records from the National Library of Australia's Trove across 14 endpoints under api.trove.nla.gov.au.

## Scope

Does not handle library lending, digitisation requests, or user-account management - use for Trove search and retrieval of digitised newspapers, gazettes, magazines, books, people, and lists only.

## Capabilities

- Run a unified search across every Trove zone through GET `/v3/result`
- Retrieve a specific work by ID through `/v3/work/{workId}`
- Look up an individual newspaper article through `/v3/newspaper/{id}`
- Look up an individual gazette article through `/v3/gazette/{id}`
- Find newspaper, gazette, or magazine titles with metadata through /titles endpoints
- Resolve a person or organisation by ID through `/v3/people/{id}`
- List Trove contributor institutions and inspect a specific contributor

## Use cases

### Historical newspaper research

Researchers and journalists can query Trove's digitised newspaper archive going back to the 19th century. GET `/v3/result` with category=newspaper and a date range returns ranked matches, and GET `/v3/newspaper/{id}` retrieves the full article including OCR text. The data is rich enough to support computational text analysis on a single topic across decades.

Example prompt: Call GET `/v3/result` with q=Federation, category=newspaper, and l-decade=190 then retrieve each article via `/v3/newspaper/{id}.`

### Genealogy and biographical lookup

Genealogy tools can resolve Australian historical persons and organisations through Trove. GET `/v3/people/{id}` returns the people record; the unified `/v3/result` search with category=people surfaces matches for a name. Combined with newspaper search, an agent can build a timeline for a historical individual from primary sources.

Example prompt: Call GET `/v3/result`?q=Henry+Lawson&category=people, then GET `/v3/people/{id}` for the top result to retrieve the full biographical record.

### Library discovery integration

A library website or chatbot can surface relevant Trove content alongside its own catalogue. `/v3/result` returns matches across all zones (books, images, music, archived websites), and `/v3/work/{workId}` returns the full work record. The API key in X-API-KEY rate-limits the integration without authenticating individual end users.

Example prompt: Embed GET `/v3/result`?q={user_query}&category=book&n=10 in a library search box and render the title, contributor, and snippet for each result.

### AI agent historical research assistant

An AI agent acting as a research assistant can search Trove, fetch full articles, and summarise results for a user without manual API calls. Through Jentic, the agent searches by intent and Jentic returns the matching Trove endpoint with the X-API-KEY attached server-side. Free Trove keys keep the integration cost-neutral while protecting per-key rate limits.

Example prompt: Search Jentic for 'search Trove', load GET `/v3/result`, and execute it with q=Eureka+Stockade&category=newspaper&n=20.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/v3/result` | Unified search across every Trove zone |
| GET | `/v3/work/{workId}` | Retrieve a single work record |
| GET | `/v3/newspaper/{id}` | Retrieve a single newspaper article with OCR text |
| GET | `/v3/newspaper/titles` | Find newspaper titles with metadata |
| GET | `/v3/gazette/{id}` | Retrieve a single gazette article |
| GET | `/v3/people/{id}` | Resolve a person or organisation record |
| GET | `/v3/list/{id}` | Retrieve a user-curated Trove list |
| GET | `/v3/contributor` | List Trove contributor institutions |

## Key resources

- **Search** — Unified Trove search across newspapers, books, magazines, images, music, people, and lists
- **Work** — Look up an individual work record by ID
- **Newspaper** — Search newspaper titles and retrieve individual articles
- **Gazette** — Search gazette titles and retrieve individual articles
- **People** — Resolve people and organisation records by ID
- **List** — Retrieve user-curated Trove lists
- **Contributor** — List contributor institutions and inspect individual contributors

## Why Jentic

- **Setup:** Wiring the Trove API by hand means adding its API key header, building requests against its host (https://api.trove.nla.gov.au), and handling its search parameters yourself. Through Jentic you install once, import the Trove API from the API Directory, store the API key once, and your agent calls it.
- **Permission scoping:** Trove is a read-only search and retrieval API, so scope by operation: limit the agent to the calls it needs, such as running a search or fetching a newspaper article, and leave out lists or contributor lookups unless you add them. You choose the operations it may call.
- **Credential handling:** Your Trove API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'search digitised newspapers' or 'fetch a work by id', and Jentic returns the matching Trove operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **OpenAlex API** — OpenAlex provides scholarly works, authors, and institutions that complement Trove's primary-source archive.
- **Wikimedia API** — Wikimedia supplies Wikipedia and Wikidata reference data that pairs with Trove primary sources.
- **NHTSA Vehicle API (vPIC)** — Another free, key-light public-sector data API - useful as a reference for similar government open-data integrations.

## FAQ

### Why is there no official OpenAPI spec for Trove API?

The National Library of Australia publishes Trove documentation but does not ship a maintained OpenAPI specification for direct download. Jentic generates and maintains this spec so that AI agents and developers can call Trove API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the Trove API use?

Trove uses an API key sent in the X-API-KEY header. Keys are free to register at trove.nla.gov.au. Through Jentic, the key is held in the vault and attached server-side, so agents never see the raw value.

### Can I retrieve the full text of a digitised newspaper article?

Yes. GET `/v3/newspaper/{id}` returns the article record including OCR-derived article text where available, ready for downstream text analysis or summarisation.

### What are the rate limits for the Trove API?

Trove enforces per-key rate limits and asks heavy users to register a project key. Exact thresholds are published at trove.nla.gov.au. If a key exceeds the limit the API returns HTTP 429 - back off and retry with a longer interval.

### How do I search Trove through Jentic?

Run pip install jentic, search Jentic for 'search Trove', and load GET `/v3/result.` Provide q, category (newspaper, book, people, image, list, magazine, music), and pagination parameters; Jentic attaches the X-API-KEY header automatically.

### Is the Trove API free?

Yes. Trove is a free public service from the National Library of Australia. There is no charge for the API key or for normal usage; key registration and Terms of Use apply.

### Can I limit what my agent is allowed to do with the Trove API?

Yes. Because you run Jentic One yourself, your own rules decide which Trove operations the agent may call and which API key it uses. Since Trove is read-only, you scope by operation: allow just the calls the agent needs, such as running a search via GET `/v3/result` or fetching an article via GET `/v3/newspaper/{id}`, while leaving out list retrieval or contributor lookups unless you explicitly add them. The Trove API key stays with your instance and is attached at execution time, so the agent can only reach the endpoints you permit.