Create, read, update and query any Salesforce object with SOQL and SOSL.
Salesforce ships in the w6w first-party pack. It declares 12 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.salesforceSalesforce runs SOQL and SOSL directly from a workflow — querying any object with SOQL, paging through large result sets with a dedicated query-more action, or searching across multiple object types at once with SOSL, and reading, creating, updating, upserting or bulk-creating records on any object by name.
Object structure is discoverable rather than assumed: every sobject in an org can be listed, and any one of them described in full — its fields, types and relationships — so a workflow can adapt to a custom object or a customized standard one instead of hardcoding field names that may not exist in a given org.
A fit for syncing any Salesforce object with an external system, running reporting queries against CRM data, or writing records into Salesforce as part of a larger process — because the actions work against any object, not a fixed set of standard ones.
Three routes to the same 12 actions. The Workflow tab is generated from Salesforce's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
limits-get Read the org's API limits and current usage. `DailyApiRequests` is the one a bulk workflow hits first.
query Run a SOQL query. Salesforce caps a page at 2000 records — follow `nextRecordsUrl` with `query-more`.
record-create Create a record of any object — Lead, Account, Contact, Opportunity, Case or a custom object.
record-create-many Create up to 200 records of one object in a single request.
record-delete Delete a record. It goes to the Recycle Bin, recoverable for about 15 days.
record-update Update a record's fields. Salesforce answers 204 with no body — use `record-get` to read it back.
record-upsert Create or update a record keyed on an External ID field. The retry-safe alternative to `record-create`.
sobject-describe Fetch an object's field metadata — API names, types, picklist values and which fields are External IDs.
sobject-list List every object in the org, standard and custom — the source of the API names other actions take.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Salesforce's real ids.
{
"manifestVersion": "2",
"name": "salesforce-example",
"steps": [
{
"id": "query",
"uses": {
"app": "io.w6w.salesforce",
"action": "query",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"soql": "<soql>"
}
}
]
}query limits-get query-more record-create record-create-many +7 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Salesforce, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Salesforce. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "record-create",
payload: {
sobject: "<value>",
fields: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action record-create --payload '{"sobject":"<value>","fields":"<value>"}' Give an AI agent Salesforce — without giving it Salesforce's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.salesforce#query",
"input": {
"soql": "<soql>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Salesforce connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Salesforce's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Salesforce itself is MIT, and the runtime that executes it is source-available (FSL).
Salesforce declares its own checks, so its health is a property of the app rather than something the host guesses at.
Salesforce Trust status for the instance this connection lives on. Per-instance, because that is the granularity an incident actually has.
Daily API request allowance and storage headroom for this org, from `/limits`. Exhausting `DailyApiRequests` locks the whole org out of the API until midnight.