Fetching Webapp Rest Api

Salesforce REST API via Data SDK Fetch

Use sdk.fetch from the Data SDK when GraphQL is not sufficient. The SDK applies authentication, CSRF handling, and base URL resolution. Always use optional chaining (sdk.fetch?.()) and handle the case where fetch is not available.

Invoke this skill when you need to call Chatter, Connect REST, Apex REST, UI API REST, or Einstein LLM endpoints.

API Version

Use the project's API version. It is typically injected as __SF_API_VERSION__; fallback to "65.0":

declare const __SF_API_VERSION__: string;
const API_VERSION = typeof __SF_API_VERSION__ !== "undefined" ? __SF_API_VERSION__ : "65.0";

Base Path

URLs are relative to the Salesforce API base. The SDK prepends the correct base path. Use paths starting with /services/....

---

Chatter API

User and collaboration data. No GraphQL equivalent.

const sdk = await createDataSDK();
const response = await sdk.fetch?.(`/services/data/v${API_VERSION}/chatter/users/me`);

if (!response?.ok) throw new Error(`HTTP ${response?.status}`);
const data = await response.json();
return { id: data.id, name: data.name };

---

Connect REST API

File and content operations.

const sdk = await createDataSDK();
const configRes = await sdk.fetch?.(`/services/data/v${API_VERSION}/connect/file/upload/config`, {
  method: "GET",
});

if (!configRes?.ok) throw new Error(`Failed to get upload config: ${configRes?.status}`);
const config = await configRes.json();
const { token, uploadUrl } = config;

---

Apex REST

Custom Apex REST resources. Requires corresponding Apex classes in the org. CSRF protection is applied automatically for services/apexrest URLs.

Example (login):

const sdk = await createDataSDK();
const response = await sdk.fetch?.("/services/apexrest/auth/login", {
  method: "POST",
  body: JSON.stringify({ email, password, startUrl: "/" }),
  headers: { "Content-Type": "application/json", Accept: "application/json" },
});

Apex REST paths do not include the API version.

---

UI API (REST)

When GraphQL cannot cover the use case. Prefer GraphQL when possible.

const sdk = await createDataSDK();
const response = await sdk.fetch?.(`/services/data/v${API_VERSION}/ui-api/records/${recordId}`);

---

Einstein LLM Gateway

AI features. Requires Einstein API setup.

const sdk = await createDataSDK();
const response = await sdk.fetch?.(`/services/data/v${API_VERSION}/einstein/llm/prompt/generations`, {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    additionalConfig: { applicationName: "PromptTemplateGenerationsInvocable" },
    promptTextorId: prompt,
  }),
});

if (!response?.ok) throw new Error(`Einstein LLM failed (${response?.status})`);
const data = await response.json();
return data?.generations?.[0]?.text ?? "";

---

General Pattern

import { createDataSDK } from "@salesforce/sdk-data";

const sdk = await createDataSDK();

if (!sdk.fetch) {
  throw new Error("Data SDK fetch is not available in this context");
}

const response = await sdk.fetch(url, {
  method: "GET", // or POST, PUT, PATCH, DELETE
  headers: { "Content-Type": "application/json", Accept: "application/json" },
  body: method !== "GET" ? JSON.stringify(payload) : undefined,
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
const data = await response.json();

---

Reference

• Parent: accessing-data — enforces Data SDK usage for all Salesforce data fetches
• GraphQL: using-graphql — use for record queries and mutations when possible
• createRecord from @salesforce/webapp-experimental/api for UI API record creation (uses SDK internally)