snow-flow 10.0.186-dev.683 → 10.0.186-dev.694

package/package.json

@@ -1,6 +1,6 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
-  "version": "10.0.186-dev.683",
+  "version": "10.0.186-dev.694",
   "name": "snow-flow",
   "description": "Snow-Flow - ServiceNow Multi-Agent Development Framework powered by AI",
   "license": "Elastic-2.0",

package/src/project/agents-template.txt

@@ -590,23 +590,23 @@ You have **direct access** to MCP tools in your environment. They are **already
 - Call the MCP tool directly once discovered
 - Tools work like built-in functions - just call them
 
-### Anti-Pattern 2: Using Scheduled Scripts / Background Scripts
+### Anti-Pattern 2: Using Script Execution for CRUD Operations
 
-**🚨 CRITICAL: `snow_schedule_script_job` and `snow_trigger_scheduled_job` are UNRELIABLE!**
+**`snow_execute_script` executes scripts reliably (~1-3s via sync REST API), but prefer dedicated tools for CRUD operations.**
 
-These tools **CANNOT execute scripts directly**. They create scheduled jobs that depend on the ServiceNow scheduler — which frequently times out, returns `executed=false`, or simply never runs the script. **Do NOT use these tools to verify, test, or validate anything.**
+`snow_execute_script` now uses synchronous REST API execution (with scheduler fallback), so it IS reliable for script execution. However, you should still prefer dedicated tools over raw GlideRecord scripts.
 
-**NEVER use `snow_schedule_script_job` or `snow_trigger_scheduled_job` for:**
-- ❌ Verifying if something was created correctly (use `snow_query_table` instead)
-- ❌ Testing GlideRecord queries (use `snow_query_table` instead)
-- ❌ Checking table/field existence (use `snow_query_table` with limit=1 instead)
-- ❌ Creating/updating artifacts (use dedicated tools!)
-- ❌ Any operation where you need reliable results
+**Prefer dedicated tools over `snow_execute_script` for:**
+- Querying records → use `snow_query_table`
+- Creating/updating artifacts → use dedicated tools (snow_deploy, snow_update, etc.)
+- Flow operations → use `snow_manage_flow`
+- Checking table/field existence → use `snow_query_table` with limit=1
 
-**What to use instead:**
-- `snow_query_table` — for querying records, checking existence, verifying data
-- `snow_manage_flow` — for flow operations
-- Dedicated tools — always prefer a specific tool over a generic script execution
+**`snow_execute_script` IS appropriate for:**
+- ✅ Custom verification scripts that need server-side logic
+- ✅ Complex operations not covered by dedicated tools
+- ✅ Debugging and testing with gs.info output
+- ✅ One-off data operations
 
 ### Anti-Pattern 3: No Mock Data, No Placeholders
 

package/src/servicenow/servicenow-automation-mcp.ts

@@ -371,7 +371,7 @@ class ServiceNowAutomationMCP {
         {
           name: "snow_schedule_script_with_output",
           description:
-            "⚠️ DEPRECATED - Use snow_schedule_script_job instead. SCHEDULES (not executes directly) a script via scheduled job. May return executed=false. ES5 only!",
+            "⚠️ DEPRECATED - Use snow_execute_script instead. SCHEDULES (not executes directly) a script via scheduled job. May return executed=false. ES5 only!",
           inputSchema: {
             type: "object",
             properties: {
@@ -400,7 +400,7 @@ class ServiceNowAutomationMCP {
         {
           name: "snow_schedule_script_sync",
           description:
-            '⚠️ DEPRECATED - Use snow_schedule_script_job instead. SCHEDULES (not sync executes) a script via scheduled job. The "sync" name is misleading - this still uses async scheduling. ES5 only!',
+            '⚠️ DEPRECATED - Use snow_execute_script instead. SCHEDULES (not sync executes) a script via scheduled job. The "sync" name is misleading - this still uses async scheduling. ES5 only!',
           inputSchema: {
             type: "object",
             properties: {

package/src/servicenow/servicenow-mcp-server.ts

@@ -220,9 +220,9 @@ class ServiceNowMCPServer {
             },
           },
           {
-            name: "snow_schedule_script_job",
+            name: "snow_execute_script",
             description:
-              "⚠️ SCHEDULES (not executes directly) server-side JavaScript via Scheduled Script Job. Creates sysauto_script + sys_trigger. May return executed=false if scheduler doesn't pick it up - check System Scheduler > Scheduled Jobs.",
+              "Execute server-side JavaScript on ServiceNow. Primary: synchronous execution via Scripted REST API (~1-3s). Fallback: scheduled job if endpoint unavailable. ES5 only (Rhino engine)!",
             inputSchema: {
               type: "object",
               properties: {
@@ -287,7 +287,7 @@ class ServiceNowMCPServer {
           case "snow_create_workflow":
             return await this.handleCreateWorkflow(args)
 
-          case "snow_schedule_script_job":
+          case "snow_execute_script":
             return await this.handleExecuteScript(args)
 
           case "snow_test_connection":

package/src/servicenow/servicenow-mcp-unified/README.md

@@ -51,7 +51,7 @@ servicenow-mcp-unified/
 │   │   ├── snow_add_uib_page_element.ts
 │   │   └── index.ts
 │   ├── automation/              # Script execution, jobs
-│   │   ├── snow_schedule_script_job.ts    # Schedule script jobs (NOT direct execution!)
+│   │   ├── snow_execute_script.ts          # Execute scripts (sync REST API + scheduler fallback)
 │   │   ├── snow_schedule_job.ts
 │   │   └── index.ts
 │   ├── advanced/                # Analytics
@@ -110,10 +110,10 @@ Tools are organized by **domain** (functional area), not by original server:
 
 ### 4. Automation Tools (automation/)
 
-**Purpose:** Script scheduling and job management
+**Purpose:** Script execution and job management
 
-- `snow_schedule_script_job` - ⚠️ SCHEDULES scripts (NOT direct execution!) via sysauto_script + sys_trigger
-- `snow_confirm_script_execution` - Confirm scripts requiring approval (also uses scheduling)
+- `snow_execute_script` - Execute scripts via synchronous REST API (~1-3s), with scheduler fallback
+- `snow_confirm_script_execution` - Confirm scripts requiring approval
 - `snow_schedule_job` - Create scheduled jobs
 - `snow_get_logs` - Access system logs
 - `snow_trace_execution` - Set up execution tracing configuration

package/src/servicenow/servicenow-mcp-unified/tools/automation/index.ts

@@ -34,11 +34,11 @@ export {
   toolDefinition as snow_discover_schedules_def,
   execute as snow_discover_schedules_exec,
 } from "./snow_discover_schedules.js"
-// Scheduled script job tool - schedules scripts via sysauto_script + sys_trigger (does NOT execute directly!)
+// Script execution tool - synchronous REST API execution with scheduler fallback
 export {
-  toolDefinition as snow_schedule_script_job_def,
-  execute as snow_schedule_script_job_exec,
-} from "./snow_schedule_script_job.js"
+  toolDefinition as snow_execute_script_def,
+  execute as snow_execute_script_exec,
+} from "./snow_execute_script.js"
 export { toolDefinition as snow_get_logs_def, execute as snow_get_logs_exec } from "./snow_get_logs.js"
 export {
   toolDefinition as snow_get_email_logs_def,

package/src/servicenow/servicenow-mcp-unified/tools/automation/snow_confirm_script_execution.ts

@@ -2,7 +2,7 @@
  * snow_confirm_script_execution - Confirm and execute approved script
  *
  * Executes a script after user approval. Only call this after user
- * explicitly approves script execution from snow_schedule_script_job with requireConfirmation=true.
+ * explicitly approves script execution from snow_execute_script with requireConfirmation=true.
  *
  * Uses sysauto_script + sys_trigger approach for reliable execution.
  *
@@ -16,7 +16,7 @@ import { createSuccessResult, createErrorResult, SnowFlowError, ErrorType } from
 export const toolDefinition: MCPToolDefinition = {
   name: "snow_confirm_script_execution",
   description:
-    "⚡ Confirms and schedules script after user approval (use after snow_schedule_script_job with requireConfirmation=true). Note: Uses same scheduled job approach.",
+    "⚡ Confirms and schedules script after user approval (use after snow_execute_script with requireConfirmation=true). Note: Uses same scheduled job approach.",
   // Metadata for tool discovery (not sent to LLM)
   category: "automation",
   subcategory: "script-execution",

package/src/servicenow/servicenow-mcp-unified/tools/automation/{snow_schedule_script_job.ts → snow_execute_script.ts}

@@ -1,5 +1,5 @@
 /**
- * snow_schedule_script_job - Execute server-side JavaScript on ServiceNow
+ * snow_execute_script - Execute server-side JavaScript on ServiceNow
  *
  * Primary: synchronous execution via Scripted REST API (~1-3s)
  * Fallback: scheduled job if endpoint unavailable
@@ -87,11 +87,11 @@ const OPERATION_SCRIPT = `(function process(request, response) {
 })(request, response);`
 
 export const toolDefinition: MCPToolDefinition = {
-  name: "snow_schedule_script_job",
+  name: "snow_execute_script",
   description:
     "Execute server-side JavaScript on ServiceNow. Primary: synchronous execution via Scripted REST API (~1-3s). Fallback: scheduled job if endpoint unavailable. Auto-deploys the executor endpoint on first use. ES5 only (Rhino engine)!",
   category: "automation",
-  subcategory: "scheduled-jobs",
+  subcategory: "script-execution",
   use_cases: ["automation", "scripts", "scheduled-jobs", "debugging", "verification"],
   complexity: "advanced",
   frequency: "high",
@@ -150,7 +150,7 @@ export const toolDefinition: MCPToolDefinition = {
 
 export async function execute(args: Record<string, unknown>, context: ServiceNowContext): Promise<ToolResult> {
   const script = args.script as string
-  const description = (args.description as string) || "Script scheduled via snow_schedule_script_job"
+  const description = (args.description as string) || "Script executed via snow_execute_script"
   const timeout = (args.timeout as number) || 30000
   const validate = args.validate_es5 !== false
   const confirmation = args.requireConfirmation === true

package/src/servicenow/servicenow-mcp-unified/tools/automation/snow_get_script_output.ts

@@ -46,7 +46,7 @@ export async function execute(args: any, context: ServiceNowContext): Promise<To
     const client = await getAuthenticatedClient(context)
 
     // Try to find the output in sys_properties
-    // snow_schedule_script_job stores output as: SNOW_FLOW_EXEC_${executionId}
+    // snow_execute_script stores output as: SNOW_FLOW_EXEC_${executionId}
     const outputMarker = `SNOW_FLOW_EXEC_${execution_id}`
 
     const outputResponse = await client.get(
@@ -62,7 +62,7 @@ export async function execute(args: any, context: ServiceNowContext): Promise<To
         await client.delete(`/api/now/table/sys_properties/${property.sys_id}`)
       }
 
-      // Organize output by level (matches snow_schedule_script_job format)
+      // Organize output by level (matches snow_execute_script format)
       const organizedOutput = {
         print: (scriptOutput.output || []).filter((o: any) => o.level === "print").map((o: any) => o.message),
         info: (scriptOutput.output || []).filter((o: any) => o.level === "info").map((o: any) => o.message),

package/src/servicenow/servicenow-mcp-unified/tools/integration/snow_rest_message_manage.ts

@@ -27,7 +27,7 @@ export const toolDefinition: MCPToolDefinition = {
 • Message-level Headers - apply to ALL methods (list_message_headers, create_message_header, delete_message_header)
 • Method-level Headers - apply to specific method (list_method_headers, create_method_header, delete_method_header)
 • Query Parameters (list_parameters, create_parameter, delete_parameter)
-• Testing (test) - generates ES5-compatible test script for snow_schedule_script_job`,
+• Testing (test) - generates ES5-compatible test script for snow_execute_script`,
   // Metadata for tool discovery (not sent to LLM)
   category: "integration",
   subcategory: "rest",
@@ -1448,11 +1448,11 @@ ${paramLines.length > 0 ? "    " + paramLines.join("\n    ") : "    // No parame
         test_params_provided: test_params,
         test_script: testScript,
         usage: {
-          description: "Use snow_schedule_script_job to execute this test script",
-          example: `snow_schedule_script_job({ script: <test_script>, description: 'Test REST: ${restMessageName}' })`,
+          description: "Use snow_execute_script to execute this test script",
+          example: `snow_execute_script({ script: <test_script>, description: 'Test REST: ${restMessageName}' })`,
           manual_test: `In ServiceNow: System Web Services > Outbound > REST Message > ${restMessageName} > ${method.name} > Test`,
         },
-        note: "Direct REST message testing requires ServiceNow script execution. Copy the test_script and run it via snow_schedule_script_job.",
+        note: "Direct REST message testing requires ServiceNow script execution. Copy the test_script and run it via snow_execute_script.",
       },
       { operation: "test_rest_message" },
     )