What the agent gets
- SOQL queries across standard and custom objects.
- Cross-object text search, for when the object is not known upfront.
- Schema discovery, so the agent can find the right objects and fields without being told.
- Related-record traversal from a parent record.
- Record creates, updates, and deletes, if you enable the write tools.
Setup
Salesforce needs two admin steps before Rippit can connect. Both are one-time. This covers production and Developer orgs; sandbox and scratch orgs use a different endpoint and are not supported here.1
Activate the SObject All server
In Salesforce, go to Setup → MCP Servers → Salesforce Servers and activate the standard SObject All server. Until this is done the endpoint returns “Server definition not found.” Allow about two minutes to propagate.
2
Create an External Client App
Salesforce does not let Rippit register itself, so create an External Client App with the
mcp_api and refresh_token scopes and PKCE enabled. Add the redirect URI Rippit shows in the connect dialog, copied exactly.3
Connect in Rippit
Paste the app’s Consumer Key into Rippit. Add the Consumer Secret only if the app has “Require Secret for Web Server Flow” checked. If that box is unchecked, leave the secret blank.
Identity and scope
The connection acts as the Salesforce user who authorized it, and enforces that user’s object and field-level security. A field the account cannot see is a field the agent cannot see. For an organization-wide connection, authorize with an account whose Salesforce permissions are appropriate for everyone who will use the agent, including scheduled runs.Tools
The six read tools are enabled by default.