Skip to main content
Glama
simonl77

Salesforce MCP Server

by simonl77

salesforce_dml_records

Insert, update, delete, or upsert Salesforce records using DML operations to manage data across objects like Accounts and Cases.

Instructions

Perform data manipulation operations on Salesforce records:

  • insert: Create new records

  • update: Modify existing records (requires Id)

  • delete: Remove records (requires Id)

  • upsert: Insert or update based on external ID field Examples: Insert new Accounts, Update Case status, Delete old records, Upsert based on custom external ID

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
operationYesType of DML operation to perform
objectNameYesAPI name of the object
recordsYesArray of records to process
externalIdFieldNoExternal ID field name for upsert operations

Implementation Reference

  • Core handler function executing DML operations (insert/update/delete/upsert) on Salesforce records, formats results with success counts and detailed error messages.
    export async function handleDMLRecords(conn: any, args: DMLArgs) {
      const { operation, objectName, records, externalIdField } = args;
    
      let result: DMLResult | DMLResult[];
      
      switch (operation) {
        case 'insert':
          result = await conn.sobject(objectName).create(records);
          break;
        case 'update':
          result = await conn.sobject(objectName).update(records);
          break;
        case 'delete':
          result = await conn.sobject(objectName).destroy(records.map(r => r.Id));
          break;
        case 'upsert':
          if (!externalIdField) {
            throw new Error('externalIdField is required for upsert operations');
          }
          result = await conn.sobject(objectName).upsert(records, externalIdField);
          break;
        default:
          throw new Error(`Unsupported operation: ${operation}`);
      }
    
      // Format DML results
      const results = Array.isArray(result) ? result : [result];
      const successCount = results.filter(r => r.success).length;
      const failureCount = results.length - successCount;
    
      let responseText = `${operation.toUpperCase()} operation completed.\n`;
      responseText += `Processed ${results.length} records:\n`;
      responseText += `- Successful: ${successCount}\n`;
      responseText += `- Failed: ${failureCount}\n\n`;
    
      if (failureCount > 0) {
        responseText += 'Errors:\n';
        results.forEach((r: DMLResult, idx: number) => {
          if (!r.success && r.errors) {
            responseText += `Record ${idx + 1}:\n`;
            if (Array.isArray(r.errors)) {
              r.errors.forEach((error) => {
                responseText += `  - ${error.message}`;
                if (error.statusCode) {
                  responseText += ` [${error.statusCode}]`;
                }
                if (error.fields && error.fields.length > 0) {
                  responseText += `\n    Fields: ${error.fields.join(', ')}`;
                }
                responseText += '\n';
              });
            } else {
              // Single error object
              const error = r.errors;
              responseText += `  - ${error.message}`;
              if (error.statusCode) {
                responseText += ` [${error.statusCode}]`;
              }
              if (error.fields) {
                const fields = Array.isArray(error.fields) ? error.fields.join(', ') : error.fields;
                responseText += `\n    Fields: ${fields}`;
              }
              responseText += '\n';
            }
          }
        });
      }
    
      return {
        content: [{
          type: "text",
          text: responseText
        }],
        isError: false,
      };
    }
  • Tool specification including name, description, and input schema defining parameters for DML operations.
    export const DML_RECORDS: Tool = {
      name: "salesforce_dml_records",
      description: `Perform data manipulation operations on Salesforce records:
      - insert: Create new records
      - update: Modify existing records (requires Id)
      - delete: Remove records (requires Id)
      - upsert: Insert or update based on external ID field
      Examples: Insert new Accounts, Update Case status, Delete old records, Upsert based on custom external ID`,
      inputSchema: {
        type: "object",
        properties: {
          operation: {
            type: "string",
            enum: ["insert", "update", "delete", "upsert"],
            description: "Type of DML operation to perform"
          },
          objectName: {
            type: "string",
            description: "API name of the object"
          },
          records: {
            type: "array",
            items: { type: "object" },
            description: "Array of records to process"
          },
          externalIdField: {
            type: "string",
            description: "External ID field name for upsert operations",
            optional: true
          }
        },
        required: ["operation", "objectName", "records"]
      }
    };
  • src/index.ts:45-63 (registration)
    Registers the salesforce_dml_records tool (via DML_RECORDS) in the listTools endpoint response.
    server.setRequestHandler(ListToolsRequestSchema, async () => ({
      tools: [
        SEARCH_OBJECTS, 
        DESCRIBE_OBJECT, 
        QUERY_RECORDS, 
        AGGREGATE_QUERY,
        DML_RECORDS,
        MANAGE_OBJECT,
        MANAGE_FIELD,
        MANAGE_FIELD_PERMISSIONS,
        SEARCH_ALL,
        READ_APEX,
        WRITE_APEX,
        READ_APEX_TRIGGER,
        WRITE_APEX_TRIGGER,
        EXECUTE_ANONYMOUS,
        MANAGE_DEBUG_LOGS
      ],
    }));
  • src/index.ts:119-131 (registration)
    Switch case dispatcher for salesforce_dml_records that validates input arguments and invokes the handleDMLRecords handler.
    case "salesforce_dml_records": {
      const dmlArgs = args as Record<string, unknown>;
      if (!dmlArgs.operation || !dmlArgs.objectName || !Array.isArray(dmlArgs.records)) {
        throw new Error('operation, objectName, and records array are required for DML');
      }
      const validatedArgs: DMLArgs = {
        operation: dmlArgs.operation as 'insert' | 'update' | 'delete' | 'upsert',
        objectName: dmlArgs.objectName as string,
        records: dmlArgs.records as Record<string, any>[],
        externalIdField: dmlArgs.externalIdField as string | undefined
      };
      return await handleDMLRecords(conn, validatedArgs);
    }
  • TypeScript interface defining the expected arguments for the DML handler, used for validation.
    export interface DMLArgs {
      operation: 'insert' | 'update' | 'delete' | 'upsert';
      objectName: string;
      records: Record<string, any>[];
      externalIdField?: string;
    }

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full behavioral burden. It adds useful context like 'requires Id' for update/delete and 'external ID field' for upsert, but doesn't disclose potential side effects (e.g., permanent deletion), permission requirements, or error handling behavior. This is a meaningful but incomplete disclosure for a mutation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with a clear introductory line, a concise bulleted list of operations, and practical examples. Every sentence contributes meaning, and the format is scannable without excessive length.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (4 parameters, no output schema, no annotations), the description covers the core operations, prerequisites, and use cases effectively. It lacks details about return values or failure modes, but for a DML tool with clear operation semantics, it provides sufficient context for an agent to select and invoke it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all parameters are documented. The description adds extra semantic value by explaining operation-specific requirements (e.g., update and delete require Id, upsert uses externalIdField), which enhances the schema's basic field descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly identifies the tool as performing DML operations (insert, update, delete, upsert) on Salesforce records, with a specific verb+resource structure. It distinguishes itself from sibling tools like salesforce_query_records and salesforce_search_objects by focusing on data manipulation rather than reading or describing.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides concrete usage context with examples for each operation (Insert new Accounts, Update Case status, Delete old records, Upsert based on custom external ID). It implicitly contrasts with read-only tools, though it doesn't explicitly state when not to use this tool or name alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.