Salesforce MCP Server (Himanshu Dwivedi)
Salesforce SOQL queries, object descriptions, and contact creation via MCP.
Data last scanned today · Reviewed today
Overview
This Salesforce MCP server integrates your CRM directly into AI environments, allowing models to query data using SOQL, retrieve object schemas, and manage contacts. By exposing specific Salesforce tools as MCP-compatible functions, it enables AI agents to perform tasks like checking org limits or running account health reviews without requiring manual interface navigation. The server features a guarded query design, restricting SOQL operations to prevent unauthorized data manipulation. It includes specific tools like 'run_soql' and 'create_contact' while maintaining a focused set of capabilities, rather than a generic or broad database interface.
Our verdict
With a security score of 22/100, this server is currently high-risk. The score reflects that the project is maintained by a single contributor, lacks an explicit software license, and has not yet undergone independent editorial review. Users should exercise caution, as the lack of a formal license and vendor support may pose compliance or long-term maintenance challenges. It is intended for developers who want to experiment with Salesforce tool integration and are prepared to configure their own restricted integration user profiles.
- Features a dedicated SOQL query guard with offline tests
- Uses minimal, narrow tools rather than generic DML access
- No external dependencies reduce the overall supply chain surface
- Includes reusable prompt templates for structured account health analysis
- Zero GitHub stars suggests limited community testing
- No software license, creating legal ambiguity for potential users
- Single-contributor project increases long-term maintenance and bus-factor risks
Tools
| Tool | Description | Risk |
|---|---|---|
| run_soql | Read-only SOQL queries | low |
| describe_object | Field names/types/labels for any SObject | low |
| create_contact | Create a Contact | medium |
| org_limits | API/storage limits snapshot | low |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| ChatGPT | ||||
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf |
Frequently asked questions
›What Salesforce credentials are required to run this server?
You need a Salesforce username, password, and security token, which are configured via environment variables.
›Does this server provide full DML access to my Salesforce org?
No, it restricts write operations to a single narrow create_contact tool instead of providing generic DML capabilities.
›How are SOQL queries protected against malicious input?
The server uses a guardrail that validates model-composed queries and blocks forbidden keywords before execution.
›What type of user account is recommended for this integration?
It is recommended to use a dedicated Integration User profile with minimal object permissions to adhere to the principle of least privilege.
›Can I test the server locally before connecting it to my client?
Yes, you can use the MCP inspector UI for local testing by running the server with mcp dev server.py.
›Does the server include any pre-configured analysis tools?
Yes, it provides an account_health_review prompt template that clients can invoke by name for data analysis.
Alternatives
Badge
Maintain this server? Add the live badge to your README.