canonical: https://jentic.com/apis/cloud.tipo.gov.tw/tipo

# Cloud Tipo TIPO Open Data API

Jentic publishes the only available OpenAPI specification for TIPO Open Data API, keeping it validated and agent-ready. The Taiwan Intellectual Property Office (TIPO) Open Data API exposes patent and trademark gazette data published by the Taiwan government, including patent search, individual patent records, gazette listings, and trademark search. It also exposes a generic open-data listing endpoint and a resource download endpoint for bulk gazette files. Some endpoints work without an API key, and others require the apikey query parameter issued by TIPO's open data portal.

## For AI agents

Search Taiwan patents and trademarks, retrieve patent gazette records, and download TIPO open-data resources by ID.

## Scope

Does not handle patent filing, prosecution, or non-Taiwan jurisdictions - use for read-only access to TIPO patent and trademark gazette data only.

## Capabilities

- Search Taiwan patent records by keyword, applicant, or patent number
- Retrieve a single patent record by its TIPO patent number
- List patent gazette publications and filter by publication period
- Search Taiwan trademark records by keyword or applicant
- Download bulk gazette resource files by resource ID
- Browse the TIPO open-data catalogue listing through a single endpoint

## Use cases

### IP Watch and Competitive Monitoring

Monitor Taiwan patent and trademark filings for a competitor or technology area by polling the search endpoints on a schedule. `/api/patent/search` and `/api/trademark/search` accept keyword and applicant filters, which is enough signal to drive a daily IP watch report for legal teams covering the Taiwan market. New filings can be deduplicated against the patent number returned by `/api/patent/{patentNo}.`

Example prompt: Call `/api/patent/search` with a competitor's company name and a publication date filter for the past 30 days, then return the new patent numbers

### Gazette Ingestion for Legal Databases

Pull TIPO patent gazette data into a commercial IP database. `/api/patent/gazette` lists gazette publications and `/api/download/{resourceId}` fetches the underlying resource file for archival, which is the standard ingestion pattern for IP analytics vendors that resell Taiwan data. The pipeline can be re-run on each gazette release to keep the downstream database current.

Example prompt: List gazette publications via `/api/patent/gazette` since the last run, then call `/api/download/{resourceId}` for each new entry

### Trademark Clearance Searches

Run a quick clearance search before filing a new trademark in Taiwan by querying `/api/trademark/search` for the proposed mark. Results return matching registered marks and pending applications, which lets a legal assistant flag conflicts before the client commits to a brand. Combined with TIPO's gazette listings, this creates a low-cost first-pass clearance tool.

Example prompt: Call `/api/trademark/search` with the proposed mark string and return any registered Taiwan trademarks that match

### Open-Data Research Aggregation

Aggregate TIPO open-data resources into a research portal alongside other government datasets. /opdata exposes the catalogue listing and `/api/download/{resourceId}` retrieves individual resource files, which is enough to wire TIPO into a multi-source open-data dashboard. This is the typical pattern for academic IP research and journalism.

Example prompt: Call /opdata to list available TIPO datasets and download the most recent patent statistics resource

### Agent-Driven IP Lookups via Jentic

AI assistants used by patent attorneys and analysts can answer 'has this idea already been filed in Taiwan?' by searching Jentic for an intent like 'search taiwan patents' and executing `/api/patent/search` behind the scenes. The TIPO API key lives in your Jentic One instance and never enters the agent's logs, which matters for law firms with strict client confidentiality controls.

Example prompt: Search Jentic for 'search taiwan patents', load `/api/patent/search`, and execute with the user's keyword

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/api/patent/search` | Search Taiwan patents |
| GET | `/api/patent/{patentNo}` | Get a patent by patent number |
| GET | `/api/patent/gazette` | List patent gazette publications |
| GET | `/api/trademark/search` | Search Taiwan trademarks |
| GET | `/api/download/{resourceId}` | Download a resource by ID |
| GET | `/opdata` | List the open-data catalogue |

## Key resources

- **patent/search** — Keyword and applicant search across Taiwan patents
- **patent/{patentNo}** — Single patent record lookup by patent number
- **patent/gazette** — Listing of patent gazette publications
- **trademark/search** — Search Taiwan trademark records
- **download/{resourceId}** — Download a bulk gazette resource file
- **opdata** — Catalogue listing for TIPO open data

## AI readiness

This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.

- **Score:** 49 / 100
- **Maturity:** Foundational
- **Dimensions:**
  - Foundational Compliance: 100 / 100
  - Developer Experience & Jentic Compatibility: 65 / 100
  - AI-Readiness & Agent Experience: 50 / 100
  - Agent Usability: 94 / 100
  - Security: 15 / 100
  - AI Discoverability: 100 / 100
- **View full report:** https://jentic.com/apis/cloud.tipo.gov.tw/tipo/scorecard
- **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/
- **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview

### Score it yourself

Every API in the directory is allowlisted, so you can re-score it with no key required.

- **Score your own API:** https://jentic.com/scorecard.md
- **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md

```sh
npx @jentic/api-scorecard-cli score <openapi-url>
```

## Why Jentic

- **Setup:** Wiring the TIPO Open Data API by hand means passing its apikey as a query parameter, reverse-engineering its patent and trademark endpoints off an open-data portal, and handling retries yourself against cloud.tipo.gov.tw/S220. Through Jentic you install once, import the TIPO Open Data API from the API Directory, store the API key once, and your agent calls it.
- **Permission scoping:** The TIPO Open Data API is read-only, with the patent number carried in the URL path (`/api/patent/{patentNo}`), so scope the agent to the operations it needs, such as searching patents or fetching a patent by number. You choose which operations are in that set, so it stays limited to the read operations you allow.
- **Credential handling:** Your TIPO API key is stored once, encrypted, by your own Jentic One instance and injected as the apikey query parameter at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'search Taiwan patents' or 'get a patent by number', and Jentic returns the matching TIPO operation with its query schema, so the agent calls the right endpoint without reverse-engineering the open-data portal.

## Related APIs

- **USPTO BDSS** — United States Patent and Trademark Office bulk data and search service.
- **Data.gov** — United States open-data catalogue across many agencies.
- **Opendatasoft** — Hosted open-data publishing platform with a uniform query API.

## FAQ

### Why is there no official OpenAPI spec for TIPO Open Data API?

Taiwan Intellectual Property Office does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call TIPO Open Data 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 TIPO Open Data API use?

TIPO uses an apiKey scheme - the key is passed as the apikey query parameter. Some endpoints accept anonymous access; others require a key issued by the TIPO open data portal. Through Jentic, the apikey is stored in the vault and injected per call so it never enters the agent's context.

### Can I search Taiwan trademarks with this API?

Yes. GET `/api/trademark/search` accepts keyword and applicant filters and returns matching Taiwan trademark records, which is enough to drive a clearance lookup or competitive monitoring report.

### What are the rate limits for the TIPO Open Data API?

TIPO does not publish a formal rate limit in this spec. Treat the service as best-effort government infrastructure - back off on 429 or 503 responses and avoid bursting more than a few requests per second per API key.

### How do I look up a Taiwan patent by number through Jentic?

Run pip install jentic, search for 'get taiwan patent by number', load the `/api/patent/{patentNo}` operation, then execute with the patent number. The call returns the gazette record for that patent.

### Can I download bulk gazette files with this API?

Yes. GET `/api/download/{resourceId}` returns the underlying gazette resource by its ID, and resource IDs come from the gazette listing at `/api/patent/gazette` or the catalogue at /opdata.

### Can I limit what my agent is allowed to do with the TIPO Open Data API?

Yes. Because you run Jentic One yourself, your own rules decide which TIPO operations the agent may call, so you can allow read-only lookups like GET `/api/patent/search` and GET `/api/patent/{patentNo}` while withholding trademark search, gazette listing, or bulk downloads via GET `/api/download/{resourceId}.` The whole API is read-only, so nothing the agent runs can create or change TIPO records. You also control the apikey credential, which your Jentic One instance injects only for the operations you have permitted.
