@sassoftware/sas-score-mcp-serverjs 1.0.1-15 → 1.0.1-17

package/.skills/sas-find-resource-strategy/SKILL.md

@@ -50,8 +50,6 @@ Generic model aliases:
 - `model jobdef` or `jobdef model` -> `jobdef`
 
 ## Mandatory rule
-
-- Use the sas-find-resource strategy to check for the existence of a resource before attempting to read or use it. This ensures that follow-up actions are based on accurate information about resource availability.
 - Do not use list tools to find a resource.
 
 ## Tool mapping
@@ -64,34 +62,47 @@ Generic model aliases:
 
 ## Strategy by resource type
 
-### Find library
+### Strategy for library lookup
+#### Mandatory inputs:
+  - Never pass a empty server name to sas-score-find-table. It must be either "cas" or "sas"
 
-If server is provided:
-- CAS: `sas-score-find-library({ name: "<lib>", server: "cas" })`
-- SAS: uppercase first, then `sas-score-find-library({ name: "<LIB>", server: "sas" })`
+#### Follow this sequence to find the library:
+- Check if the library is one of the default libraries in CAS or SAS Server.  If found, report that as found with the associated server. 
 
-If server is not provided:
-1. If library location is already known from earlier conversation or if the library is one of the default libraries, reuse that known server and do not re-check.
-2. Check CAS first - `sas-score-find-library({ name: "<lib>", server: "cas" })`
-3. If not found, check SAS with uppercase library name - `sas-score-find-library({ name: "<LIB>", server: "sas" })`
-4. Report where found, or report not found in either.
+- If server is provided:
+  1. CAS: `sas-score-find-library({ name: "<lib>", server: "cas" })`
+  2. SAS: uppercase lib, then `sas-score-find-library({ name: "<LIB>", server: "sas" })`
+
+If server is not provided: 
+1. Check CAS first - `sas-score-find-library({ name: "<lib>", server: "cas" })`
+2. If not found, uppercase lib and then check SAS server - `sas-score-find-library({ name: "<LIB>", server: "sas" })`
+3. Report where found, or report not found in either.
+
+### Strategy to find a table
 
-### Find table
+#### Mandatory inputs:
+- Never pass a empty server name to sas-score-find-table. It must be either "cas" or "sas"
 
 Required inputs:
 - library name
 - table name
 
 If server is provided:
-- CAS: `sas-score-find-table({ lib: "<lib>", name: "<table>", server: "cas" })`
-- SAS: uppercase library first, then `sas-score-find-table({ lib: "<LIB>", name: "<table>", server: "sas" })`
+- CAS: `sas-score-find-table({ lib: "<library>", name: "<table>", server: "cas" })`
+- SAS: 
+  - uppercase library 
+  - use `sas-score-find-table({ lib: "<LIBRARY>", name: "<table>", server: "sas" })`
 
 If server is not provided:
 1. If table location is already known from earlier conversation context, reuse that known server and do not re-check.
-2. If `lib.table` uses a known default CAS library, check CAS first.
-3. If `lib.table` uses a known default SAS library, check SAS first (uppercase library).
-4. If library is unknown (not in known defaults), check CAS first, then SAS (uppercase library).
-5. If found in both, report both and ask user which server to use for follow-up reads.
+2. If the library is one of known defaults, use the associated server (CAS or SAS). Make sure to uppercase the library name for SAS server lookup.
+
+3. If library is not one of the known defaults, find the table using the following sequence
+  1.  check in CAS server - `sas-score-find-table({ lib: "<library>", name: "<table>", server: "cas" })`
+  2.  if not found,
+   - uppercase library
+   -  check in SAS server - `sas-score-find-table({ lib: "<LIBRARY>", name: "<table>", server: "sas" })`  
+ 3. If not found, report that the table was not found in either server.
 
 ### Find models by type
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sassoftware/sas-score-mcp-serverjs",
-  "version": "1.0.1-15",
+  "version": "1.0.1-17",
   "description": "A mcp server for SAS Viya",
   "author": "Deva Kumar <deva.kumar@sas.com>",
   "license": "Apache-2.0",

package/src/toolSet/modelScore.js

@@ -18,7 +18,7 @@ DO NOT USE for: find model, model metadata, list models, run programs/jobs, quer
 PARAMETERS
 - model: string — model name (required, exact match)
 - scenario: object — input data as JSON (optional, defaults to {}). Example: {age:45, income:60000}
-- uflag: boolean (default: false) — prefix model fields with underscore when true
+- uflag: boolean (optional, defaults to false) — prefix model fields with underscore when true
 
 ROUTING RULES
 - "score with model X using a=1, b=2" → { model: "X", scenario: {a:1, b:2}, uflag: false }
@@ -45,7 +45,7 @@ Returns predictions, probabilities, scores merged with input data. Returns error
     inputSchema: z.object({
       model: z.string(),
       scenario: z.union([z.record(z.any()), z.string()]).optional(),
-      uflag: z.boolean().optional()
+      uflag: z.boolean()
     }),
       
     handler: async (iparams) => {