@sassoftware/sas-score-mcp-serverjs 1.0.1-3 → 1.0.1-4

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sassoftware/sas-score-mcp-serverjs",
-  "version": "1.0.1-3",
+  "version": "1.0.1-4",
   "description": "A mcp server for SAS Viya",
   "author": "Deva Kumar <deva.kumar@sas.com>",
   "license": "Apache-2.0",
@@ -42,8 +42,7 @@
     "openApi.json",
     "openApi.yaml",
     "scripts",
-    ".skills_github",
-    ".skills_claude"
+    ".skills"
   ],
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.29.0",

package/src/oauthHandlers/callback.js

@@ -75,7 +75,7 @@ async function callback(req, res, pkceStore, codeStore, appContext) {
     // which was part of the payload from the client to /oauth/authorize
     // we trust since it was associated with the valid PKCE state
     console.error("[Note] OAuth callback complete, redirecting to MCP client");
-    console.log(pending.clientRedirectUri.toString())
+    console.error(pending.clientRedirectUri.toString())
     return res.redirect(`${pending.clientRedirectUri}?${redirectParams}`);
   } catch (err) {
     console.error("[Error] OAuth callback handler error:", err);

package/src/processHeaders.js

@@ -34,7 +34,7 @@ function processHeaders(req, res, next, cache, appContext) {
   let token = (hdr != null) ? hdr.slice(7) : null;
   //console.error("[Note] Authorization token", token);
   debugger;
-  console.log('>>>',appContext.AUTHFLOW);
+  console.error('>>>',appContext.AUTHFLOW);
   if (appContext.AUTHFLOW === 'bearer') {
     debugger;
     let startAuth = false;

package/src/setupSkills.js

@@ -40,7 +40,7 @@ function setupSkills(clientName) {
         const entries = fs.readdirSync(dir, { withFileTypes: true });
 
         for (const entry of entries) {
-            console.log(indent + entry.name);
+            console.error(indent + entry.name);
 
             if (entry.isDirectory()) {
                 listExpandedFolder(path.join(dir, entry.name), indent + "  ");

package/.skills_claude/README.md

@@ -1,303 +0,0 @@
-# Claude Desktop — SAS Agent Setup & Deployment Guide
-
-This guide explains how to set up the SAS agent in Claude Desktop using the `.claude` folder.
-
-## What's Included
-
-The `.claude` folder contains:
-- ✅ Complete SAS agent configuration (same as `.github/` but organized for Claude)
-- ✅ All 6 specialized SAS skills
-- ✅ MCP server configuration template
-- ✅ System prompt for Claude Desktop
-- ✅ Setup and usage documentation
-
-## Prerequisites
-
-Before you start:
-- ✅ Claude Desktop installed (latest version recommended)
-- ✅ Node.js installed and in your PATH
-- ✅ SAS Viya environment with OAuth credentials
-- ✅ Access to your SAS Viya server from your machine
-- ✅ (Optional) CA certificate file if using self-signed TLS
-
-## Setup Steps
-
-### Step 1: Gather Your SAS Viya Credentials
-
-You'll need:
-- **VIYA_URL** — Your SAS Viya server URL (e.g., `https://sas-viya.example.com`)
-- **VIYA_CLIENT_ID** — OAuth client ID
-- **VIYA_CLIENT_SECRET** — OAuth client secret
-- **VIYA_TLS_CERT** — (Optional) Path to CA certificate file
-
-### Step 2: Configure the MCP Server Connection
-
-1. Open `.claude/claude-desktop-config.json`
-2. Update the configuration with your credentials:
-
-```json
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": [
-        "/absolute/path/to/7-skill-integration/cli.js"
-      ],
-      "env": {
-        "VIYA_URL": "https://your-sas-viya-url.com",
-        "VIYA_CLIENT_ID": "your-client-id",
-        "VIYA_CLIENT_SECRET": "your-client-secret",
-        "VIYA_TLS_CERT": "/path/to/ca-cert.pem"
-      }
-    }
-  }
-}
-```
-
-**Important notes:**
-- Use **absolute paths** for both the CLI path and certificate
-- Ensure the path to `cli.js` is correct for your system
-- Include the leading `/` for Unix/Mac paths; for Windows, use `C:\path\to\...`
-
-### Step 3: Place Configuration in Claude Desktop Directory
-
-Copy the updated `claude_desktop_config.json` to your Claude Desktop config location:
-
-**macOS/Linux:**
-```bash
-cp /path/to/7-skill-integration/.claude/claude-desktop-config.json ~/.claude/claude_desktop_config.json
-```
-
-**Windows:**
-```powershell
-Copy-Item "C:\path\to\7-skill-integration\.claude\claude-desktop-config.json" "$env:APPDATA\Claude\claude_desktop_config.json"
-```
-
-**Note:** If you have other MCP servers configured, merge the `mcpServers` sections instead of replacing the entire file.
-
-### Step 4: Add System Prompt to Claude
-
-1. Open **Claude Desktop**
-2. Go to **Settings** → **Custom Instructions**
-3. Open `.claude/claude-desktop-system-prompt.md`
-4. Copy the entire contents
-5. Paste into the custom instructions field in Claude
-6. Save
-
-### Step 5: Restart Claude Desktop
-
-Restart Claude Desktop to activate the new configuration and system prompt.
-
-## Verification
-
-After setup, verify the connection works:
-
-1. Start a chat with Claude
-2. Ask: **"list jobs"**
-3. You should see available SAS jobs from your Viya environment
-4. If successful, the MCP connection is working ✅
-
-## Configuration Details
-
-### MCP Server Configuration (claude-desktop-config.json)
-
-The configuration tells Claude Desktop how to:
-- Start the SAS MCP server (using Node.js)
-- Connect to your SAS Viya environment
-- Pass credentials securely
-
-**Key fields:**
-- `command` — The Node.js interpreter
-- `args` — Path to the MCP server CLI
-- `env` — Environment variables for authentication and TLS
-
-### System Prompt (claude-desktop-system-prompt.md)
-
-The system prompt instructs Claude to:
-- Treat itself as a SAS domain specialist
-- Load skills before using MCP tools
-- Classify ambiguous SAS terms
-- Verify assets before execution
-- Never hallucinate SAS resources
-
-## Using the SAS Agent
-
-Once configured, you can start using the agent:
-
-### Basic Commands
-
-```
-You: "Find library Public"
-Agent: Loads sas-find-library-smart skill, checks CAS and SAS, returns result
-
-You: "List tables in Public"
-Agent: Loads sas-list-tables-smart skill, enumerates tables
-
-You: "Score customers in Public.customers with model churn.mas"
-Agent: Loads sas-read-and-score + sas-score-workflow, returns predictions
-```
-
-### Available Skills
-
-See `skills-index.md` for quick reference, or each skill's file for details:
-
-1. **sas-request-classifier** — Disambiguate SAS terms
-2. **sas-find-library-smart** — Find CAS or SAS libraries
-3. **sas-list-tables-smart** — List tables in a library
-4. **sas-read-strategy** — Choose right read tool (raw vs. analytical)
-5. **sas-read-and-score** — Read table data + score with model
-6. **sas-score-workflow** — Route scoring by model type (.job, .mas, etc.)
-
-## Troubleshooting
-
-### "MCP Server Not Found"
-
-**Symptom:** Claude says the MCP server can't be found.
-
-**Solutions:**
-1. Verify the path to `cli.js` is absolute and correct
-2. Check that Node.js is installed: `node --version`
-3. Ensure Node.js is in your PATH: `which node` (Mac/Linux) or `where node` (Windows)
-4. Restart Claude Desktop
-5. Check the file permissions on `cli.js`
-
-### "Authentication Failed"
-
-**Symptom:** Claude connects but SAS operations fail with auth errors.
-
-**Solutions:**
-1. Verify `VIYA_CLIENT_ID` and `VIYA_CLIENT_SECRET` are correct
-2. Ensure credentials are still valid (check if they've expired)
-3. Verify `VIYA_URL` is accessible from your machine: `curl https://your-viya-url`
-4. Check network connectivity and firewall rules
-5. If using a proxy, configure it in your environment
-
-### "TLS Certificate Error"
-
-**Symptom:** SSL/certificate verification fails.
-
-**Solutions:**
-1. Verify the path in `VIYA_TLS_CERT` is correct and absolute
-2. Ensure the certificate file is readable
-3. If using self-signed cert, try `VIYA_TLS_SKIP_VERIFY=true` (development only, not recommended for production)
-4. Request the correct CA certificate from your SAS administrator
-
-### "No Tools Available"
-
-**Symptom:** Claude says no SAS tools are available.
-
-**Solutions:**
-1. Restart Claude Desktop
-2. Check the MCP server logs for errors
-3. Verify the SAS Viya server is running and accessible
-4. Try a simple test: ask "list jobs"
-5. Check your custom instructions are loaded correctly
-
-### "Claude Doesn't Load Skills"
-
-**Symptom:** Claude doesn't automatically load the right skill for your request.
-
-**Solutions:**
-1. Explicitly reference the skill: "Use sas-find-library-smart to find library Public"
-2. Check that `claude-desktop-system-prompt.md` is loaded in custom instructions
-3. Restart Claude Desktop to reload the system prompt
-4. Verify the skill files exist in `.claude/skills/`
-
-## Advanced Configuration
-
-### Using Multiple MCP Servers
-
-If you have other MCP servers configured:
-
-```json
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": ["/path/to/cli.js"],
-      "env": { ... }
-    },
-    "other-server": {
-      "command": "...",
-      "args": [...],
-      "env": { ... }
-    }
-  }
-}
-```
-
-### Environment Variables
-
-You can also set environment variables system-wide instead of in the config:
-
-```bash
-export VIYA_URL="https://your-sas-viya.com"
-export VIYA_CLIENT_ID="your-client-id"
-export VIYA_CLIENT_SECRET="your-client-secret"
-export VIYA_TLS_CERT="/path/to/ca-cert.pem"
-```
-
-Then simplify the config to just reference `cli.js`.
-
-### Debugging
-
-To debug MCP server issues, check:
-1. Claude Desktop logs: `~/Library/Logs/Claude Desktop` (Mac) or `%APPDATA%\Claude\logs` (Windows)
-2. MCP server output: Check if there are error messages from Node.js
-3. Test the CLI directly: `node /path/to/cli.js`
-
-## File Structure Reference
-
-```
-.claude/
-├── copilot-instructions.md              ← Core SAS instructions
-├── claude-desktop-system-prompt.md      ← Claude Desktop prompt
-├── claude-desktop-config.json           ← MCP config (customize this)
-├── README.md                             ← This file
-├── instructions.md                      ← Usage instructions
-├── skills-index.md                      ← Skills quick reference
-├── agents/
-│   └── sas-viya-scoring-expert.md       ← Agent definition
-└── skills/
-    ├── sas-request-classifier/SKILL.md
-    ├── sas-find-library-smart/SKILL.md
-    ├── sas-list-tables-smart/SKILL.md
-    ├── sas-read-and-score/SKILL.md
-    ├── sas-read-strategy/SKILL.md
-    └── sas-score-workflow/SKILL.md
-```
-
-## Deployment
-
-### Local Development
-
-For local development:
-1. Keep the `.claude` folder in the repository
-2. Update `claude-desktop-config.json` locally (don't commit credentials)
-3. Share setup steps with team members
-
-### Production Deployment
-
-For production use:
-1. Store credentials in environment variables, not in config file
-2. Use a secrets management system for OAuth credentials
-3. Commit only the `.claude` template without sensitive data
-4. Have users configure their own credentials locally
-
-## Support & Documentation
-
-- **Setup issues?** See "Troubleshooting" above
-- **Want to use a skill?** Open `.claude/skills-index.md`
-- **Need usage guidance?** See `.claude/instructions.md`
-- **SAS domain details?** See `.claude/copilot-instructions.md`
-- **Skill specifics?** Open the skill's file directly: `.claude/skills/<name>/SKILL.md`
-
-## Next Steps
-
-1. ✅ Complete the setup steps above
-2. ✅ Verify the connection ("list jobs" test)
-3. ✅ Read `instructions.md` for how to use the agent
-4. ✅ Try a simple task: "Find library Public"
-5. ✅ Explore skills: See `skills-index.md`
-
-Happy SAS Viya scoring! 🎯

package/.skills_claude/TESTING_GUIDE.md

@@ -1,252 +0,0 @@
-# Testing Guide: Local Claude Desktop Setup
-
-This guide explains how to test the SAS agent locally by placing configuration in your home directory (`~/.claude/`).
-
-## Quick Start
-
-### 1. Create the `.claude` directory in your home folder
-
-```bash
-mkdir -p ~/.claude
-```
-
-### 2. Copy the configuration file
-
-```bash
-cp /path/to/7-skill-integration/.claude/claude-desktop-config.json ~/.claude/
-```
-
-### 3. Edit the configuration with your credentials
-
-```bash
-nano ~/.claude/claude-desktop-config.json
-```
-
-Update the following fields with your SAS Viya environment details:
-
-```json
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": [
-        "/absolute/path/to/7-skill-integration/cli.js"
-      ],
-      "env": {
-        "VIYA_URL": "https://your-sas-viya.com",
-        "VIYA_CLIENT_ID": "your-client-id",
-        "VIYA_CLIENT_SECRET": "your-client-secret",
-        "VIYA_TLS_CERT": "/path/to/ca-cert.pem"
-      }
-    }
-  }
-}
-```
-
-**Important:**
-- Use the **absolute path** to `cli.js` in your repository
-- Include the leading `/` for Unix/Mac paths; for Windows, use `C:\path\to\...`
-
-### 4. Add the system prompt to Claude Desktop
-
-1. Open **Claude Desktop**
-2. Go to **Settings** → **Custom Instructions**
-3. Open `.claude/claude-desktop-system-prompt.md` from the repository
-4. Copy the entire contents
-5. Paste into the custom instructions field
-6. Save
-
-### 5. Restart Claude Desktop
-
-Close and reopen Claude Desktop to load the new configuration and system prompt.
-
-### 6. Test the connection
-
-In a chat with Claude, ask:
-
-```
-list jobs
-```
-
-You should see a list of available SAS jobs from your Viya environment. If successful, the MCP connection is working! ✅
-
----
-
-## What Files Go Where?
-
-### Files that must go in `~/.claude/`:
-- ✅ `claude-desktop-config.json` — **Required** for MCP server connection
-
-### Files that don't need to be copied:
-- ❌ Skills files — They live in the repository and are referenced in documentation
-- ❌ Agent definitions — Documentation only
-- ❌ Skill markdown files — For reference, not needed for execution
-
-### System prompt:
-- **Option A** — Copy `claude-desktop-system-prompt.md` contents to Claude Desktop Settings → Custom Instructions (recommended)
-- **Option B** — Keep it in the repository and reference it
-
----
-
-## Directory Structure After Setup
-
-Your home directory will look like this:
-
-```
-~/
-├── .claude/
-│   └── claude-desktop-config.json    ← Your configuration
-├── ... (other home directory files)
-```
-
-The repository stays intact:
-
-```
-/path/to/7-skill-integration/
-├── .claude/
-│   ├── README.md
-│   ├── instructions.md
-│   ├── skills-index.md
-│   ├── copilot-instructions.md
-│   ├── claude-desktop-system-prompt.md
-│   ├── claude-desktop-config.json    ← Template (not modified)
-│   ├── agents/
-│   └── skills/
-├── cli.js
-├── ... (other repo files)
-```
-
----
-
-## How It Works
-
-1. Claude Desktop reads `~/.claude/claude-desktop-config.json`
-2. It starts the MCP server (Node.js process running `cli.js`)
-3. The MCP server connects to your SAS Viya environment using the credentials you provided
-4. Claude loads the system prompt from custom instructions
-5. You can now use the SAS agent with all skills available
-
-The repository's `.claude/` folder serves as the **template and documentation** — it doesn't need to be in your home directory.
-
----
-
-## Troubleshooting
-
-### "MCP Server Not Found"
-
-**Check:**
-1. Verify the path to `cli.js` is absolute and correct
-2. Test Node.js is available: `node --version`
-3. Verify the file exists: `ls -la /path/to/cli.js`
-4. Restart Claude Desktop
-
-**Solution:**
-```bash
-# Verify the path works
-node /absolute/path/to/7-skill-integration/cli.js --help
-```
-
-### "Authentication Failed"
-
-**Check:**
-1. Verify credentials in `~/.claude/claude-desktop-config.json`
-2. Ensure SAS Viya is accessible: `curl https://your-viya-url`
-3. Verify credentials are still valid
-
-### "No SAS Tools Available"
-
-**Check:**
-1. System prompt is loaded in Claude Custom Instructions
-2. Restart Claude Desktop after adding the system prompt
-3. Try a simple test: "list jobs"
-
----
-
-## Testing Different Configurations
-
-### Test with a local development environment:
-
-```json
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": ["/Users/yourname/projects/7-skill-integration/cli.js"],
-      "env": {
-        "VIYA_URL": "http://localhost:7480",
-        "VIYA_CLIENT_ID": "dev-client",
-        "VIYA_CLIENT_SECRET": "dev-secret"
-      }
-    }
-  }
-}
-```
-
-### Test with production environment:
-
-```json
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": ["/opt/7-skill-integration/cli.js"],
-      "env": {
-        "VIYA_URL": "https://sas-viya-prod.company.com",
-        "VIYA_CLIENT_ID": "prod-client-id",
-        "VIYA_CLIENT_SECRET": "prod-client-secret",
-        "VIYA_TLS_CERT": "/etc/ssl/certs/company-ca.pem"
-      }
-    }
-  }
-}
-```
-
----
-
-## Next Steps After Testing
-
-### If testing is successful:
-1. Document your setup
-2. Share the setup steps with your team (without credentials)
-3. Consider using environment variables instead of config file for production
-4. Document any team-specific customizations
-
-### If testing fails:
-1. Check troubleshooting section above
-2. Review MCP server logs
-3. Verify SAS Viya environment is accessible
-4. Check credentials are correct and current
-
----
-
-## Important Notes
-
-### Security
-
-- ⚠️ **Never commit credentials** to version control
-- ⚠️ **Never share** your `~/.claude/claude-desktop-config.json` file
-- ✅ Use environment variables for sensitive data when possible
-- ✅ Use a secrets management system for team deployments
-
-### Paths
-
-- ✅ Use absolute paths (starting with `/` or `C:\`)
-- ❌ Don't use relative paths like `./cli.js`
-- ❌ Don't use `~` in the config file (Claude Desktop may not expand it correctly)
-
-### File Permissions
-
-Ensure the config file is readable:
-
-```bash
-chmod 600 ~/.claude/claude-desktop-config.json
-```
-
----
-
-## Support
-
-- **Setup questions?** See the main `README.md` in `.claude/`
-- **How to use the agent?** See `instructions.md` in `.claude/`
-- **Which skill to use?** See `skills-index.md` in `.claude/`
-- **Domain guidance?** See `copilot-instructions.md` in `.claude/`

package/.skills_claude/agents/sas-viya-scoring-expert.md

@@ -1,58 +0,0 @@
----
-name: SAS Viya Scoring Expert
-description: Specialized SAS and Viya agent that classifies requests, selects the right SAS skill, and uses MCP tools safely for jobs, CAS data, libraries, models, scoring, and content workflows.
----
-
-# SAS Viya Scoring Expert
-
-You are a SAS Viya expert agent.
-
-Your job is to help users work with SAS and Viya resources through the SAS MCP server.
-Treat requests as domain-specific SAS tasks, not generic coding tasks.
-
-## Default behavior
-Before using MCP tools:
-- Determine whether the request is about jobs, code, CAS data, libraries, models, scoring, content, or environment issues.
-- If the request includes ambiguous terms such as model, score, scoring, read, query, job, code, table, content, asset, or resource, classify the request before acting.
-- Prefer loading the most relevant SAS skill before using low-level tools.
-- If confidence is low, ask one focused clarifying question.
-- Prefer discovery and inspection before execution, publish, scoring, deploy, write, or destructive actions.
-
-## Skill-first policy
-Use skills as the primary source of SAS workflow guidance.
-Load one or more relevant SAS skills before using tools when the request is ambiguous, cross-domain, or execution-oriented.
-Do not load unrelated skills.
-
-## Routing policy
-When a request is ambiguous or could map to more than one SAS domain:
-- Start with classification.
-- Identify the most likely SAS asset or workflow type.
-- Choose the best matching SAS skill.
-- Only then select MCP tools.
-
-## Ambiguity policy
-These terms are overloaded in SAS and Viya workflows and should not be interpreted casually:
-- model
-- score
-- scoring
-- read
-- query
-- job
-- code
-- table
-- content
-- asset
-- resource
-
-If the meaning is unclear, ask one targeted clarifying question or use discovery-oriented skills before any execution step.
-
-## Tool usage policy
-- Prefer read-only discovery before execution.
-- Confirm the target asset type before running jobs, scoring data, publishing models, or modifying content.
-- If tool results contradict the initial interpretation, correct course explicitly and continue.
-- Never invent asset names, identifiers, libraries, or model types.
-
-## Response style
-Be concise, explicit, and domain-aware.
-State which SAS concept or asset type you are acting on when ambiguity is possible.
-Prefer short structured answers when guiding the user.

package/.skills_claude/claude-desktop-config.json

@@ -1,16 +0,0 @@
-{
-  "mcpServers": {
-    "sas-mcp-server": {
-      "command": "node",
-      "args": [
-        "/path/to/7-skill-integration/cli.js"
-      ],
-      "env": {
-        "VIYA_URL": "https://your-sas-viya-url.com",
-        "VIYA_CLIENT_ID": "your-client-id",
-        "VIYA_CLIENT_SECRET": "your-client-secret",
-        "VIYA_TLS_CERT": "/path/to/ca-cert.pem"
-      }
-    }
-  }
-}