2.1

Author: stevevillardi

.github/workflows/publish.yml

@@ -0,0 +1,43 @@
+name: Publish to npm
+
+on:
+  release:
+    types: [published]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 22.x
+          cache: 'npm'
+      - run: npm ci
+      - run: npm run build
+      - run: npm test -- tests/unit/
+      - run: npm test -- tests/smoke.test.ts
+
+  publish:
+    needs: test
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      id-token: write
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 22.x
+          cache: 'npm'
+          registry-url: 'https://registry.npmjs.org'
+      - run: npm ci
+      - name: Set version from release tag
+        run: |
+          TAG_VERSION="${GITHUB_REF_NAME#v}"
+          echo "Publishing version $TAG_VERSION from tag $GITHUB_REF_NAME"
+          npm version "$TAG_VERSION" --no-git-tag-version --allow-same-version
+      - run: npm run build
+      - run: npm publish --access public
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}

.gitignore

@@ -41,6 +41,9 @@ Thumbs.db
 coverage/
 .nyc_output/
 
+# LLM eval reports
+tests/eval/reports/
+
 # PM2
 .pm2/
 CLAUDE.md

README.md

@@ -10,7 +10,7 @@ A Model Context Protocol (MCP) server that provides secure access to the LogicMo
 
 ## Features
 
-- **Comprehensive Resource Management**: Devices, device groups, websites, website groups, collectors, alerts, users, dashboards, collector groups, and device metrics
+- **Comprehensive Resource Management**: Devices, device groups, websites, website groups, collectors, alerts, users, dashboards, collector groups, device metrics, SDTs, and OpsNotes
 - **Device Metrics & Data**: Retrieve monitoring data including datasources, instances, and time-series metrics
 - **Batch Operations**: Process multiple items efficiently with rate limiting and error handling
 - **Guided Workflows**: Built-in prompts for complex tasks like exporting device metrics
@@ -42,7 +42,7 @@ npm install -g logicmonitor-api-mcp
 
 ```bash
 # Clone the repository
-git clone https://github.com/stevevillardi/logicmonitor-api-mcp.git
+git clone https://github.com/LogicMonitor/logicmonitor-api-mcp.git
 cd logicmonitor-api-mcp
 
 # Install dependencies
@@ -102,11 +102,12 @@ HTTP mode allows remote access and is suitable for shared deployments:
 
 1. **Start the server:**
 ```bash
-# With environment variables
-LM_ACCOUNT=your-account LM_BEARER_TOKEN=your-bearer-token PORT=3000 logicmonitor-api-mcp
+# With environment variables (default port is 3000)
+LM_ACCOUNT=your-account LM_BEARER_TOKEN=your-bearer-token logicmonitor-api-mcp
 
 # Or use a .env file
-echo "PORT=3000" > .env
+cp .env.example .env
+# Edit .env with your credentials
 logicmonitor-api-mcp
 ```
 
@@ -269,17 +270,35 @@ Retrieve device monitoring data including datasources, instances, and time-serie
 Key features:
 - Wildcard filtering for datasources (e.g., `datasourceIncludeFilter: "CPU*"`)
 - Batch instance data retrieval (e.g., all CPU cores at once)
-- Flexible time ranges (ISO 8601 dates or Unix epochs, defaults to last 24 hours)
+- Flexible time ranges (relative strings like `-6h`/`-7d`, ISO 8601 dates, Unix epochs, or `"now"` — defaults to last 24 hours)
 - Formatted output with timestamps and metric values
 
+#### `lm_sdt`
+Manage Scheduled Down Times (maintenance windows):
+- **list** - List SDTs with filtering (e.g., `isEffective:true` for active SDTs)
+- **get** - Get SDT details by ID (string format, e.g., `"R_42"`)
+- **create** - Create one-time or recurring SDTs targeting devices, groups, websites, collectors, or datasources
+- **update** - Update SDT schedule, comment, or target
+- **delete** - Delete/end SDTs (single or batch)
+
+SDT types: ResourceSDT (device), ResourceGroupSDT, WebsiteSDT, WebsiteGroupSDT, CollectorSDT, DeviceDataSourceSDT, and more.
+
+#### `lm_opsnote`
+Manage operational notes for change tracking:
+- **list** - List ops notes with filtering (by tags, createdBy, happenedOn, etc.)
+- **get** - Get note details by ID
+- **create** - Create notes with optional scopes (device, service, deviceGroup, serviceGroup) and tags
+- **update** - Update note text, scopes, or tags
+- **delete** - Delete notes (single or batch)
+
 #### `lm_session`
 Manage session context and variables using standard CRUD operations:
 
 **Operations:**
 - **list** - Get session history (recent tool calls)
   - Parameters: `limit` (optional, 1-50)
 - **get** - Get session context or specific variable
-  - Parameters: `key` (optional - if omitted, returns full context), `historyLimit`, `includeResults`
+  - Parameters: `key` (optional - if omitted, returns full context), `fields` (project specific fields from arrays), `index` (single item by position), `limit` (first N items), `historyLimit`, `includeResults`
 - **create** - Store a new session variable
   - Parameters: `key` (required), `value` (required)
   - Use for storing results for batch operations with applyToPrevious
@@ -329,6 +348,26 @@ A comprehensive workflow that guides you through exporting monitoring data from
 
 The prompt will guide the AI through each step, ensuring proper data collection and formatting.
 
+### `batch-device-update`
+A guided workflow for finding devices by filter, reviewing them, and applying updates in bulk.
+
+**Arguments:**
+- `device_filter` (required) - Filter to select devices (e.g., `"displayName:*prod*"`, `"hostStatus:dead"`)
+- `update_description` (required) - Description of what to update (e.g., `"disable alerting"`, `"add custom property env=production"`)
+
+**Workflow Steps:**
+1. List devices matching the filter
+2. Review matched devices for confirmation
+3. Apply batch update via `applyToPrevious`
+4. Verify results
+
+### `alert-triage`
+A guided workflow for listing, filtering, and triaging LogicMonitor alerts.
+
+**Arguments:**
+- `severity_filter` (optional) - Alert severity to focus on: `"critical"`, `"error"`, `"warning"`, or `"all"` (default)
+- `resource_filter` (optional) - Filter to scope alerts to specific resources (e.g., `"monitorObjectName:*prod*"`)
+
 ## Key Features
 
 ### Automatic ID Resolution
@@ -444,6 +483,7 @@ npm run dev
 
 # Build
 npm run build
+```
 
 ### Debug Mode
 
@@ -458,7 +498,7 @@ LOG_LEVEL=debug logicmonitor-api-mcp
 - **Session Management**: Stateful connections with cleanup
 - **Rate Limiting**: Automatic retry with exponential backoff
 - **Batch Processing**: Concurrent operations with partial failure handling
-- **Input Validation**: Joi schemas ensure data integrity
+- **Input Validation**: Zod schemas ensure data integrity
 
 ## Contributing
 

THIRD-PARTY-NOTICES.md

@@ -19,13 +19,13 @@ The following table summarizes the licenses used by this project's dependencies:
 
 | Package | Version | License | Repository |
 |---------|---------|---------|------------|
-| @modelcontextprotocol/sdk | 1.20.2 | MIT | https://github.com/modelcontextprotocol/typescript-sdk |
-| axios | 1.13.0 | MIT | https://github.com/axios/axios |
-| dotenv | 16.6.1 | BSD-2-Clause | https://github.com/motdotla/dotenv |
-| express | 5.1.0 | MIT | https://github.com/expressjs/express |
-| helmet | 8.1.0 | MIT | https://github.com/helmetjs/helmet |
-| joi | 17.13.3 | BSD-3-Clause | https://github.com/hapijs/joi |
-| winston | 3.18.3 | MIT | https://github.com/winstonjs/winston |
+| @modelcontextprotocol/sdk | ^1.27.1 | MIT | https://github.com/modelcontextprotocol/typescript-sdk |
+| axios | ^1.13.6 | MIT | https://github.com/axios/axios |
+| dotenv | ^17.3.1 | BSD-2-Clause | https://github.com/motdotla/dotenv |
+| express | ^5.2.1 | MIT | https://github.com/expressjs/express |
+| express-rate-limit | ^8.3.1 | MIT | https://github.com/express-rate-limit/express-rate-limit |
+| winston | ^3.19.0 | MIT | https://github.com/winstonjs/winston |
+| zod | ^4.3.6 | MIT | https://github.com/colinhacks/zod |
 
 ## License Texts
 
@@ -108,5 +108,5 @@ https://github.com/LogicMonitor/logicmonitor-api-mcp/issues
 
 ---
 
-*Last updated: November 14, 2025*
+*Last updated: March 21, 2026*
 

ecosystem.config.js

@@ -39,10 +39,10 @@ export default {
   coverageReporters: ['text', 'lcov', 'html'],
   coverageThreshold: {
     global: {
-      branches: 50,
-      functions: 50,
-      lines: 50,
-      statements: 50,
+      branches: 70,
+      functions: 70,
+      lines: 70,
+      statements: 70,
     },
   },
   testTimeout: 30000,