- FIELD MAP
- DATA TYPES
- PICKLISTS
- HUMAN REVIEW
Short answer
How can I audit my Salesforce schema without using the Setup menu?
Your AI uses the Connector to pull object lists and field definitions into your chat. You get a full map of custom fields, types, and picklist values. Here is the breakdown of what you can do.
The audit outcomes
What you get from a schema audit
The AI pulls raw metadata and organizes it into these specific views.
FIELD MAP
Custom field inventory
The AI identifies every custom field in a specific object. You get a clean list of names and API names.
DATA TYPES
Type and length validation
The AI checks if fields are text, numbers, or booleans. This helps you spot data mismatches.
PICKLISTS
Value set verification
The AI pulls all valid picklist options for a field. You can quickly see if your dropdowns match your requirements.
HUMAN REVIEW
Logic and dependency checks
The AI flags the data, but you decide if the schema logic actually makes sense for your business processes.
The workflow
What your AI does when the request arrives.
The AI moves from broad discovery to specific field details.
Discover objects
The AI scans your environment to find the specific objects you want to inspect.
sf_list_objectsInspect field details
Once you pick an object, the AI pulls the full metadata for every field inside it.
sf_describe_objectMap data types
The AI reads the technical definitions to categorize each field's type and constraints.
sf_describe_objectSummarize schema
The AI compiles the raw metadata into a readable report or a structured list for you.
sf_describe_objectsf_list_objects
Try it
Copy these to start.
Use these prompts to begin your audit.
Starting points
Swap out the object names like 'Account' or 'Custom_Object__c' with your actual Salesforce names.
Salesforce Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_dXvoArPaPUkxjgPIyVRNAViK8yMiulYVKtQj5F9V/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Salesforce capabilities are ready to use.
{
"mcpServers": {
"salesforce-admin-metadata-mcp": {
"url": "https://edge.vinkius.com/vk_preview_dXvoArPaPUkxjgPIyVRNAViK8yMiulYVKtQj5F9V/mcp"
}
}
}List all the custom objects in my Salesforce org.
Show me all the fields on the Account object, including their data types.
What are the picklist values for the 'Industry' field on the Lead object?
Find all custom text fields on the Opportunity object that have a length greater than 50.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Salesforce Admin & Metadata once, then ask.
Just link your account once. Your credentials stay encrypted, and you can immediately start asking questions about your schema in your preferred AI client.
Connect Salesforce Admin & Metadata to your AIFAQ
How this task behaves
- 01
Can the AI change my field settings?
No. The AI can only read metadata. It cannot create, delete, or modify any fields or objects in Salesforce.
- 02
Does the AI see my actual record data?
No. This Connector only accesses the schema and metadata. It does not read the individual rows or values stored in your records.
- 03
Can I audit permission sets this way?
No. These tools are strictly for object and field metadata. They do not provide access to user permissions or security settings.
- 04
How does it know which objects are custom?
The AI reads the metadata properties provided by the tools to distinguish between standard and custom objects.
- 05
Will it work with Sandbox environments?
Yes, as long as you connect the Connector to the specific Salesforce instance you want to audit.
More questions about Salesforce Admin & Metadata? The Connector page answers them. See everything the Salesforce Admin & Metadata Connector can do
Connect Salesforce to Claude, Cursor, ChatGPT & more
