canonical: https://jentic.com/apis/naver.com/naver

# NAVER Open API

Jentic publishes the only available OpenAPI specification for NAVER Open API, keeping it validated and agent-ready. NAVER Open API is the umbrella interface to NAVER's Korean platform services - search across blogs, news, books, encyclopedia, images, shopping, and local listings, plus Papago neural machine translation, language detection, Korean romanisation, URL shortening, geocoding, and NAVER Login (OAuth 2.0) with user profile retrieval. Authentication uses paired X-Naver-Client-Id and X-Naver-Client-Secret headers for most endpoints, with a bearer token flow for the Login profile lookup.

## For AI agents

Run NAVER searches across blogs, news, shopping, and images, translate text with Papago, geocode Korean addresses, and retrieve NAVER Login user profiles.

## Scope

Does not handle NAVER advertising, NAVER Cloud Platform infrastructure, or NAVER Pay transactions - use for NAVER search, Papago translation, NAVER maps, URL shortening, and NAVER Login only.

## Capabilities

- Search NAVER blogs, news articles, books, encyclopedia entries, images, shopping products, and local businesses
- Translate text with Papago neural machine translation across NAVER-supported language pairs
- Detect the language of a text snippet via the Papago detectLangs endpoint
- Romanise Korean text into Latin script for display and accessibility use cases
- Geocode Korean addresses to coordinates and reverse-geocode coordinates back to addresses
- Shorten URLs using the NAVER URL shortener service
- Authenticate users via NAVER Login OAuth 2.0 and retrieve the resulting user profile

## Use cases

### Korean Market Search Aggregation

Product and content teams targeting Korea often need to surface what NAVER users see for a query across blogs, news, and shopping. The `/v1/search/blog.json`, `/v1/search/news.json`, and `/v1/search/shop.json` endpoints return ranked NAVER results that drop straight into a market-research dashboard. A handful of calls cover the dominant search verticals on the Korean web in seconds.

Example prompt: Call `/v1/search/news.json` with query='반도체' and display=10 to return the top 10 recent NAVER news results.

### Papago Translation in Localisation Workflows

Localisation teams shipping Korean and other Asian-language content use Papago for high-quality NMT between Korean, English, Chinese, and Japanese. POST to `/v1/papago/n2mt` with the source and target language codes returns translated text that can be folded into a CMS pipeline. `/v1/papago/detectLangs` is the natural pre-step when the source language of user-generated content is unknown.

Example prompt: POST to `/v1/papago/n2mt` with source=ko, target=en, and the user's input text, and return the translated string.

### Korean Address Geocoding for Maps

Map and logistics products serving Korea need to convert local addresses into coordinates and back. The `/v1/map/geocode` endpoint accepts a Korean street address and returns latitude and longitude, while `/v1/map/reversegeocode` resolves coordinates back to an administrative address. Combining both operations covers the round-trip needs of route planning and store-locator features.

Example prompt: Call `/v1/map/geocode` with query='서울특별시 강남구 테헤란로 152' and return the latitude and longitude pair.

### NAVER Login for Korean User Onboarding

Apps targeting Korean users widely support NAVER Login as a primary social sign-in. The `/oauth2.0/authorize` and `/oauth2.0/token` endpoints implement the standard OAuth 2.0 authorisation-code flow, and `/v1/nid/me` returns the authenticated user's profile fields. Wiring these three calls in sequence delivers a working NAVER Login implementation in a day.

Example prompt: After the user returns from `/oauth2.0/authorize`, exchange the code at `/oauth2.0/token`, then call `/v1/nid/me` with the bearer token to fetch the user's email and nickname.

### Agent-Driven NAVER Access via Jentic

Agents helping users research the Korean web or process Korean content need a single, reliable entry point to NAVER's surface area. Through Jentic the agent searches by intent (for example, translating Korean text), Jentic returns the relevant NAVER operation, and the agent executes it with the client headers stored in your Jentic One instance. The agent never handles raw NAVER credentials.

Example prompt: Search Jentic for 'translate Korean text with Papago', load the schema for `/v1/papago/n2mt`, and execute it on the user's input string.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/v1/search/news.json` | Search NAVER news articles |
| GET | `/v1/search/blog.json` | Search NAVER blogs |
| GET | `/v1/search/shop.json` | Search NAVER shopping listings |
| POST | `/v1/papago/n2mt` | Translate text with Papago |
| POST | `/v1/papago/detectLangs` | Detect language of input text |
| GET | `/v1/map/geocode` | Geocode a Korean address |
| GET | `/v1/map/reversegeocode` | Reverse-geocode coordinates |
| GET | `/v1/nid/me` | Retrieve the authenticated NAVER Login user profile |

## Key resources

- **Search** — Search across NAVER blogs, news, books, encyclopedia, images, shopping, and local results
- **Papago** — Translate text, detect language, and romanise Korean strings
- **Maps** — Geocode and reverse-geocode Korean addresses
- **URL Shortener** — Shorten URLs through the NAVER shortener
- **NAVER Login** — OAuth 2.0 authorisation and user profile retrieval

## Why Jentic

- **Setup:** Wiring NAVER by hand means learning its X-Naver-Client-Id and X-Naver-Client-Secret headers, adding a bearer token for NAVER Login, splitting traffic across the openapi.naver.com and nid.naver.com hosts, and handling errors yourself. Through Jentic you install once, import the NAVER Open API from the API Directory, store the client credentials once, and your agent calls it.
- **Permission scoping:** NAVER carries its search terms and translation text in the query and request body rather than in the URL path, so scope the agent to the operations it needs, such as news search and Papago translation. You choose the operations it may call, so geocoding or the NAVER Login profile fetch are not included unless you add them.
- **Credential handling:** Your NAVER X-Naver-Client-Id and X-Naver-Client-Secret, and any NAVER Login access token, are stored once, encrypted, by your own Jentic One instance and injected at execution time. They never enter the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'translate Korean text with Papago' or 'search NAVER news for a keyword', and Jentic returns the matching operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Kakao API** — Kakao's developer platform offers competing search, maps, and login services for the Korean market.
- **Google Cloud Translation API** — Global neural machine translation service with broader language coverage than Papago.
- **Google Custom Search JSON API** — Google search results that complement NAVER for cross-region search aggregation.
- **Bing Search API** — Microsoft's web search API, used as an alternative when NAVER coverage is not required.

## FAQ

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

NAVER does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call NAVER Open 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 NAVER Open API use?

Most NAVER Open API endpoints require a paired client ID and client secret sent as the X-Naver-Client-Id and X-Naver-Client-Secret request headers. NAVER Login profile reads at `/v1/nid/me` use a bearer access token obtained from the OAuth 2.0 flow at `/oauth2.0/authorize` and `/oauth2.0/token.`

### Can I translate text with the NAVER Open API?

Yes. POST to `/v1/papago/n2mt` with source and target language codes plus the input string and Papago returns the translated text. `/v1/papago/detectLangs` identifies the source language when it is unknown, and `/v1/papago/romanization` romanises Korean text.

### What are the rate limits for the NAVER Open API?

NAVER applies daily call quotas per service per application - for example the search endpoints are typically limited to 25,000 calls per day per app. The OpenAPI spec does not include the exact figures; check the developer console at developers.naver.com for your application's quota.

### How do I run a NAVER search through Jentic?

Run pip install jentic, search Jentic for 'search NAVER news for a keyword', load the schema for `/v1/search/news.json`, and execute it with the query parameter. Jentic injects the X-Naver-Client-Id and X-Naver-Client-Secret headers from the vault. Run it through Jentic One, the self-hosted execution layer.

### Does the NAVER Open API support geocoding outside Korea?

No. The `/v1/map/geocode` and `/v1/map/reversegeocode` endpoints are intended for Korean addresses and coordinate systems. Use a global geocoder for international coverage.

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

Yes. Because you run Jentic One yourself, your own rules decide which NAVER operations the agent may call, so you can allow just the ones it needs, such as news search at `/v1/search/news.json` and Papago translation at `/v1/papago/n2mt.` Operations you do not grant, like address geocoding at `/v1/map/geocode` or the NAVER Login profile fetch at `/v1/nid/me`, stay off limits unless you add them. Your X-Naver-Client-Id and X-Naver-Client-Secret, and any NAVER Login access token, are stored by your own instance and injected at execution time, never entering the agent's prompt or logs.