Linkup
Search result sources, sourced answers, and structured outputs remain distinct modes.
On this page
Normalized search
Normalized search selects a source-result output with fast depth and no image results. It maps returned source names, URLs, and content. Sourced answers and structured output belong on the native surface.
import { createSearch } from '@belweave/search-sdk';
import { linkup } from '@belweave/search-sdk/linkup';
const apiKey = process.env.LINKUP_API_KEY;
if (!apiKey) throw new Error('LINKUP_API_KEY is required');
const client = createSearch({ provider: linkup({ apiKey }) });
const response = await client.web.search({
query: 'SQLite write concurrency',
limit: 5,
});
console.log(response.results, response.attempts);Depth and outputType change work and response semantics. Do not pass a sourced answer to code expecting source rows. The current native reference uses a JSON body with q, depth, and outputType; older overview examples may differ.
100 include / 50 exclude under SDK policy.
Provider-native client
import { createLinkupClient } from '@belweave/search-sdk/native-linkup';
const apiKey = process.env.LINKUP_API_KEY;
if (!apiKey) throw new Error('LINKUP_API_KEY is required');
const client = createLinkupClient({ apiKey, timeoutMs: 10_000 });
const response = await client.search({ body: { q: 'SQLite write concurrency', depth: 'fast', outputType: 'searchResults' } });
console.log(response.data, response.metadata);Read provider data from response.data and sanitized transport details from response.metadata. Native TypeScript response types are not runtime schema validation. Preserve the operation's native response and source identities.
Native operation inventory
Methods available on this native client. Paths are relative to each operation's fixed base; a provider may use more than one origin. Eligible operations retry only when you opt into a client retry policy.
| Method | HTTP / relative path | Response / retry eligibility |
|---|---|---|
balance | GET /v1/credits/balance | json · eligible |
createExtract | POST /v1/extract | json · no automatic retry |
listExtract | GET /v1/extract | json · eligible |
getExtract | GET /v1/extract/{id} | json · eligible |
fetchPage | POST /v1/fetch | json · no automatic retry |
search | POST /v1/search | json · no automatic retry |
createTasks | POST /v1/tasks | json · no automatic retry |
listTasks | GET /v1/tasks | json · eligible |
getTask | GET /v1/tasks/{id} | json · eligible |
responses | POST /v1/responses | json · no automatic retry |
createResearch | POST /v1/research | json · no automatic retry |
listResearch | GET /v1/research | json · eligible |
getResearch | GET /v1/research/{id} | json · eligible |
rewindSearch | POST /v1/rewind/search | json · no automatic retry |
rewindFetch | POST /v1/rewind/fetch | json · no automatic retry |
Limits and operation lifecycle
- Pass
signaland per-calltimeoutMson native request input when needed. - Persist a returned job identifier before explicitly reading or polling the same job. Native clients do not automatically poll.
- SSE data is incremental provider event text. Parse and validate terminal or error events according to the provider contract; there is no reconnect or replay.
- Native operations may create, delete, ingest, or incur charges. A successful HTTP response is not proof that a remote job completed.
Official sources
Use the official reference for current schemas, account access, and operation-specific requirements.