secondbrainos-mcp-server 1.2.2 → 1.2.4

package/README.md

@@ -1,11 +1,11 @@
 # Second Brain OS MCP Server
 
-This package provides a Model Context Protocol (MCP) server for integrating Second Brain OS with Claude Desktop.
+This package provides a Model Context Protocol (MCP) server for integrating Second Brain OS with Claude Desktop and Claude Code.
 
 ## Prerequisites
 
 - Node.js v16 or higher
-- Claude Desktop application
+- Claude Desktop application and/or Claude Code CLI
 - A valid Second Brain OS account with the Claude MCP feature enabled
 
 ## Installation
@@ -28,6 +28,16 @@ This will:
 3. Create the necessary configuration files for Claude Desktop
 4. Configure the server to dynamically fetch your available API endpoints
 
+### Claude Code
+
+To add the server to Claude Code:
+
+```bash
+claude mcp add secondbrainos-mcp-server -- npx secondbrainos-mcp-server
+```
+
+Or add it to your project's `.mcp.json` file.
+
 ## Features
 
 ### Enhanced OpenAPI Support
@@ -41,13 +51,23 @@ The server uses `@samchon/openapi` library for robust OpenAPI handling, providin
 - Automatic request formatting based on OpenAPI specifications
 - Support for complex parameters and nested objects
 
-### Prompts (Workflows)
-The server exposes your Second Brain OS workflows as MCP prompts, making them available in Claude Desktop's attach menu:
+### Prompts
+The server exposes three types of MCP prompts, available in the client's attach menu:
+
+#### Skills (Workflows)
 - Automatically discovers your workflows via the `runPromptChain` service
-- Each workflow appears as a selectable prompt with its name and description
-- Selecting a prompt fetches the full prompt chain (ordered instructions) and injects them into the conversation
+- Each skill appears with a `[Skill]` prefix and returns a structured document with `skill_id`, `name`, `description`, and an ordered list of prompts (metadata only, no instructions)
 - Supports an optional `user_input` argument to provide additional context
 
+#### Agents
+- Discovers your AI agents via the `getAIAgentsSchema` service
+- Each agent appears with an `[Agent]` prefix and returns a structured document containing `agent_id`, `name`, `description`, `behaviour_and_instructions`, `searchMyKnowledge_collection_id`, `actions` (with id, name, description, body_parameters), and `workflows` (with enriched prompt metadata)
+- Supports an optional `user_input` argument
+
+#### Knowledge Bases
+- Aggregates all `searchMyKnowledge_collection_id` values from your agents
+- Appears as a single `[Knowledge Bases]` prompt returning an array of collection IDs
+
 ### Better Error Handling
 - Detailed error messages for debugging
 - Proper handling of authentication failures, bad requests, and service errors
@@ -124,14 +144,14 @@ The server requires the following environment variables:
 1. **Schema Fetching**: On startup, the server fetches your personalized OpenAPI schema from Second Brain OS
 2. **Schema Conversion**: The `@samchon/openapi` library converts the schema to an optimized format for LLM function calling
 3. **MCP Tools**: Each API endpoint becomes an MCP tool that Claude can use
-4. **MCP Prompts**: If the `runPromptChain` service is available, your workflows are exposed as selectable prompts in the client UI
+4. **MCP Prompts**: Your skills (workflows), AI agents, and knowledge bases are exposed as selectable prompts in the client UI
 5. **Function Execution**: When Claude calls a tool, the server executes the corresponding API call with proper authentication
 
 ## Troubleshooting
 
-### Claude Desktop doesn't see the server
-1. Ensure Claude Desktop is completely quit before running the setup
-2. Check the configuration file was created at the correct location
+### Client doesn't see the server
+1. **Claude Desktop**: Ensure it's completely quit before running setup. Check the configuration file was created at the correct location.
+2. **Claude Code**: Run `claude mcp add` or check your `.mcp.json` configuration.
 3. Review the logs at the platform-specific location mentioned during setup
 
 ### Authentication errors

package/build/index.js

@@ -248,12 +248,70 @@ class SecondBrainOSServer {
                     console.error('Failed to fetch agents for prompts/list:', error);
                 }
             }
+            // Add Knowledge Bases prompt (collects searchMyKnowledge_collection_ids from agents)
+            if (this.getAIAgentsSchemaPath) {
+                try {
+                    const agents = await this.fetchAndEnrichAgents();
+                    const collectionIds = agents
+                        .map((a) => a.searchMyKnowledge_collection_id)
+                        .filter((id) => id && id.length > 0);
+                    if (collectionIds.length > 0) {
+                        prompts.push({
+                            name: "Knowledge Bases",
+                            title: "[Knowledge Bases]",
+                            description: "Knowledge base collection IDs available to agents",
+                            arguments: [
+                                {
+                                    name: "user_input",
+                                    description: "Optional context or input",
+                                    required: false
+                                }
+                            ]
+                        });
+                    }
+                }
+                catch (error) {
+                    console.error('Failed to build knowledge bases prompt:', error);
+                }
+            }
             return { prompts };
         });
         // Get a specific prompt (workflow prompt chain or agent document)
         this.server.setRequestHandler(GetPromptRequestSchema, async (request) => {
             const promptName = request.params.name;
             const userInput = request.params.arguments?.user_input;
+            // Check if this is the Knowledge Bases prompt
+            if (promptName === "Knowledge Bases") {
+                const agents = await this.fetchAndEnrichAgents();
+                const collectionIds = agents
+                    .map((a) => a.searchMyKnowledge_collection_id)
+                    .filter((id) => id && id.length > 0);
+                const knowledgeBasesDocument = {
+                    searchMyKnowledge_collection_ids: collectionIds
+                };
+                const messages = [
+                    {
+                        role: "user",
+                        content: {
+                            type: "text",
+                            text: JSON.stringify(knowledgeBasesDocument, null, 2)
+                        }
+                    }
+                ];
+                if (userInput) {
+                    messages.push({
+                        role: "user",
+                        content: {
+                            type: "text",
+                            text: userInput
+                        }
+                    });
+                }
+                return {
+                    description: `Knowledge Bases (${collectionIds.length} collection${collectionIds.length !== 1 ? 's' : ''})`,
+                    messages
+                };
+            }
             // Check if this is an agent prompt
             const agentId = this.agentNameToId.get(promptName);
             if (agentId) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "secondbrainos-mcp-server",
-  "version": "1.2.2",
+  "version": "1.2.4",
   "description": "Second Brain OS MCP Server for Claude Desktop",
   "type": "module",
   "main": "build/index.js",