@ctera/n8n-nodes-ctera-ai 0.1.0 → 0.1.2

package/LICENSE

@@ -21,3 +21,4 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 SOFTWARE.
 
 
+

package/README.md

@@ -1,238 +1,58 @@
 # n8n-nodes-ctera-ai
 
-This is an n8n community node that provides integration with CTERA AI MCP (Model Context Protocol) server for intelligent document search and retrieval.
+n8n community node for CTERA AI MCP integration - enables semantic search, chat, and expert discovery.
 
-[n8n](https://n8n.io/) is a [fair-code licensed](https://docs.n8n.io/reference/license/) workflow automation platform.
+[n8n](https://n8n.io/) is a workflow automation platform.
 
 ## Installation
 
-### Community Installation (Recommended)
-
-For n8n version 0.187.0 and above, you can install this node directly from npm:
-
-#### Standard Installation
-
-```bash
-npm install -g @ctera/n8n-nodes-ctera-ai
-```
-
-Then restart your n8n instance.
-
-#### Docker Installation
-
-For Docker-based n8n installations:
-
 ```bash
-# Install the node inside the running container
-docker exec -it <container-name> npm install -g @ctera/n8n-nodes-ctera-ai
-
-# Restart the container
-docker restart <container-name>
+cd ~/.n8n/nodes
+npm install @ctera/n8n-nodes-ctera-ai
 ```
 
-Replace `<container-name>` with your actual container name (e.g., `n8n`, `n8n_n8n_1`).
-
-**Note**: The node will persist across container restarts but will need to be reinstalled if the container is recreated. To make it permanent, consider creating a custom Docker image or using volume mounts.
-
-### Manual Installation (Development)
-
-1. Navigate to your n8n installation:
-   ```bash
-   cd ~/.n8n/nodes
-   ```
-
-2. Clone the repository:
-   ```bash
-   git clone https://github.com/ctera/ctera-n8n-nodes.git
-   cd ctera-n8n-nodes
-   ```
-
-3. Install dependencies and build:
-   ```bash
-   npm install
-   npm run build
-   ```
-
-4. Restart n8n
-
-For more details, see the [n8n community nodes installation guide](https://docs.n8n.io/integrations/community-nodes/installation/).
+Then restart n8n. The **CTERA Data Intelligence** node will appear in your node list.
 
 ## Operations
 
-The CTERA AI node supports the following operations:
-
-### List Experts
-Get a list of all available experts (AI knowledge bases) in the system.
-
-**No parameters required**
-
-**Output**: Array of expert objects containing:
-- `id`: Expert ID for API calls
-- `name`: Display name of the expert
-- `description`: Brief description of the expert's knowledge domain
-- `mcp_endpoint`: Dedicated MCP endpoint URL for this expert
-
-### Semantic Search
-Search an expert's knowledge base using advanced semantic search.
-
-**Parameters**:
-- **Expert** (required): Select from available experts (dropdown populated via List Experts)
-- **Query** (required): Your search question or topic
-- **Additional Options**:
-  - Modified After: Filter files modified on or after this date (YYYY-MM-DD)
-  - Modified Before: Filter files modified before this date (YYYY-MM-DD)
-  - Max Results: Number of relevant chunks to return (1-50, default: 5)
-
-**Output**: Array of relevant document chunks with:
-- Text content
-- Source file paths
-- Relevance scores
-- Metadata (dates, file types, etc.)
-
-### Chat
-Have a conversational interaction with an AI expert about their knowledge domain.
-
-**Parameters**:
-- **Expert** (required): Select from available experts
-- **Message** (required): Your message or question to the expert
-- **Conversation ID** (optional): Continue an existing conversation thread
-
-**Output**: 
-- AI expert's response
-- Conversation ID (for follow-up messages)
-- Citations to source documents (if applicable)
+- **List Experts** - Get all available AI experts (knowledge bases)
+- **Semantic Search** - Search within an expert's knowledge base
+- **Chat** - Have a conversation with an AI expert
 
 ## Credentials
 
-This node requires CTERA AI MCP API credentials:
-
-### Setup Instructions
-
-1. **Generate a Bearer Token**:
-   ```bash
-   # Replace YOUR_MCP_SERVER, YOUR_OID, and email with your actual values
-   curl -k "https://YOUR_MCP_SERVER/api/tokens/generate?oid=YOUR_OID&email=your.email@example.com" | jq -r .token
-   ```
+**1. Log into Admin UI** with your SSO credentials
 
-2. **Add Credentials in n8n**:
-   - Go to **Credentials** → **New Credential**
-   - Search for "CTERA AI MCP API"
-   - Fill in:
-     - **MCP Server URL**: Your MCP server base URL (e.g., `https://mcp.example.com`)
-     - **Bearer Token**: Paste the token from step 1
-     - **Ignore SSL Issues**: Enable if using self-signed certificates
+**2. Generate MCP Bearer Token:**
 
-3. **Test the Credentials**: Click "Test" to verify connection
-
-## Example Workflows
-
-### Competitive Intelligence Daily Report
-
-```
-Schedule Trigger (Daily 9 AM)
-  ↓
-CTERA AI: List Experts
-  ↓
-Split In Batches
-  ↓
-CTERA AI: Semantic Search
-  - Expert: {{$json.id}}
-  - Query: "Recent updates, announcements, and new features"
-  - Modified After: {{$today.minus(7, 'days')}}
-  ↓
-Aggregate Results
-  ↓
-Send Email: Daily Intelligence Summary
-```
-
-### Interactive Q&A Chatbot
-
-```
-Webhook (Receive Question)
-  ↓
-CTERA AI: Chat
-  - Expert: (pre-configured expert ID)
-  - Message: {{$json.question}}
-  - Conversation ID: {{$json.conversationId}}
-  ↓
-Format Response
-  ↓
-Return Webhook Response
-```
-
-### Document Discovery Workflow
-
-```
-Schedule Trigger (Hourly)
-  ↓
-CTERA AI: Semantic Search
-  - Expert: "Product Documentation"
-  - Query: "Installation and setup guides"
-  - Modified After: {{$today.minus(1, 'hour')}}
-  ↓
-Filter: If results not empty
-  ↓
-Send Slack Notification
-```
-
-## Compatibility
-
-Tested with:
-- n8n version: 1.0.0+
-- CTERA AI MCP Server: 1.6.16+
-
-## Resources
-
-* [n8n community nodes documentation](https://docs.n8n.io/integrations/community-nodes/)
-* [CTERA AI MCP Architecture](https://cteranet.atlassian.net/wiki/spaces/AI/pages/4757454995)
-* [Model Context Protocol](https://modelcontextprotocol.io/)
-
-## Development
-
-### Build
-
-```bash
-npm run build
-```
-
-### Watch Mode
+Navigate to the MCP Tokens section in Admin UI, or use the API:
 
 ```bash
-npm run dev
+curl -k -X POST "https://YOUR_ADMIN_URL/admin/api/mcp-tokens/generate" \
+  -H "Cookie: connect.sid=YOUR_SESSION_COOKIE" \
+  -H "Content-Type: application/json" \
+  -d '{"days": 30}' | jq -r .token
 ```
 
-### Linting
+**3. Configure in n8n:**
 
-```bash
-npm run lint
-npm run lintfix
-```
+- **MCP Server URL**: `https://mcp.your-domain.com`
+- **Bearer Token**: Paste the token from step 2
+- **Ignore SSL Issues**: Enable for self-signed certificates
 
-### Publishing
+## Example Use Cases
 
-For maintainers publishing new versions to npm, see [PUBLISH.md](PUBLISH.md) for detailed instructions.
+- **Daily Intelligence Reports** - Schedule searches across experts for recent updates
+- **Q&A Chatbots** - Build webhook-based chatbots using the Chat operation
+- **Document Discovery** - Monitor for new content and send notifications
 
-## Version History
+## Resources
 
-### 0.1.0
-- Initial public release
-- List Experts operation  
-- Semantic Search operation
-- Chat with Expert operation
-- Bearer token authentication
+- [n8n Community Nodes](https://docs.n8n.io/integrations/community-nodes/)
+- [GitHub Issues](https://github.com/ctera/ctera-n8n-nodes/issues)
 
 ## License
 
-[MIT](LICENSE)
-
-## Support
-
-For issues and questions:
-- GitHub Issues: https://github.com/ctera/ctera-n8n-nodes/issues
-- CTERA Support: support@ctera.com
-
-
-
-
+[MIT](LICENSE) - See [PUBLISH.md](PUBLISH.md) for publishing guidelines.
 
 

package/dist/credentials/CteraAiMcpApi.credentials.js

@@ -5,7 +5,7 @@ class CteraAiMcpApi {
     constructor() {
         this.name = 'cteraAiMcpApi';
         this.displayName = 'CTERA AI MCP API';
-        this.documentationUrl = 'https://cteranet.atlassian.net/wiki/spaces/AI/pages/4757454995';
+        this.documentationUrl = 'https://github.com/ctera/ctera-n8n-nodes#readme';
         this.properties = [
             {
                 displayName: 'MCP Server URL',

package/dist/nodes/CteraAi/CteraAi.node.js

@@ -150,7 +150,6 @@ class CteraAi {
                 if (operation === 'listExperts') {
                     toolName = 'ctera_list_experts';
                     serverUrl = `${mcpServerUrl}/mcp-bearer`;
-                    // No parameters needed
                 }
                 else if (operation === 'search') {
                     toolName = 'expert_semantic_search';
@@ -200,19 +199,20 @@ class CteraAi {
                         'Content-Type': 'application/json',
                         'Authorization': `Bearer ${bearerToken}`,
                     },
-                    body: requestBody,
-                    json: true,
+                    body: JSON.stringify(requestBody),
                     skipSslCertificateValidation: allowUnauthorizedCerts,
                 });
+                // Parse JSON response
+                const parsedResponse = typeof response === 'string' ? JSON.parse(response) : response;
                 // Handle MCP error responses
-                if (response.error) {
-                    throw new n8n_workflow_1.NodeOperationError(this.getNode(), `MCP Error: ${response.error.message || JSON.stringify(response.error)}`, { itemIndex: i });
+                if (parsedResponse.error) {
+                    throw new n8n_workflow_1.NodeOperationError(this.getNode(), `MCP Error: ${parsedResponse.error.message || JSON.stringify(parsedResponse.error)}`, { itemIndex: i });
                 }
                 // Parse the result from MCP response
                 let result;
-                if (response.result && response.result.content) {
+                if (parsedResponse.result && parsedResponse.result.content) {
                     // MCP returns content array
-                    const content = response.result.content[0];
+                    const content = parsedResponse.result.content[0];
                     if (content && content.type === 'text') {
                         try {
                             // Try to parse as JSON if it looks like JSON
@@ -228,14 +228,14 @@ class CteraAi {
                     }
                 }
                 else {
-                    result = response.result;
+                    result = parsedResponse.result;
                 }
                 returnData.push({
                     json: {
+                        ...items[i].json,
                         operation,
                         toolName,
                         result,
-                        ...items[i].json,
                     },
                     pairedItem: { item: i },
                 });
@@ -244,8 +244,8 @@ class CteraAi {
                 if (this.continueOnFail()) {
                     returnData.push({
                         json: {
-                            error: error.message,
                             ...items[i].json,
+                            error: error.message,
                         },
                         pairedItem: { item: i },
                     });

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ctera/n8n-nodes-ctera-ai",
-  "version": "0.1.0",
+  "version": "0.1.2",
   "description": "n8n community node for CTERA AI MCP integration - enables semantic search, chat, and expert discovery",
   "keywords": [
     "n8n-community-node-package",