refactor!: rename to @codemcp/knowledge with ade-knowledge binary

.github/workflows/pr.yml

@@ -15,6 +15,8 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/action-setup@v4
+        with:
+          version: 10.32.1
 
       - name: Setup Node.js
         uses: actions/setup-node@v5

.github/workflows/release.yml

@@ -19,6 +19,8 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/action-setup@v4
+        with:
+          version: 10.32.1
 
       - name: Setup Node.js
         uses: actions/setup-node@v5
@@ -67,6 +69,8 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/action-setup@v4
+        with:
+          version: 10.32.1
 
       - name: Setup Node.js
         uses: actions/setup-node@v5

.vibe/docs/requirements.md

@@ -183,7 +183,7 @@ Complex requirements for unwanted behavior also include the If-Then keywords.
 
 - The system SHALL be distributable as NPM package
 - The system SHALL provide executable binary for MCP server
-- WHEN installed globally THEN the system SHALL be available as "agentic-knowledge-mcp" command
+- WHEN installed globally THEN the system SHALL be available as "npx @codemcp/knowledge" command
 - The system SHALL follow semantic versioning for releases
 - The system SHALL include TypeScript type definitions for programmatic use
 
@@ -206,7 +206,7 @@ Complex requirements for unwanted behavior also include the If-Then keywords.
 
 **Acceptance Criteria:**
 
-- WHEN CLI command "agentic-knowledge init {docset-id}" executed THEN the system SHALL clone configured Git repositories
+- WHEN CLI command "npx @codemcp/knowledge init {docset-id}" executed THEN the system SHALL clone configured Git repositories
 - WHEN Git repository requires authentication THEN the system SHALL use system Git credentials
 - WHEN Git repository URL is invalid THEN the system SHALL return clear error message
 - WHEN Git repository is unreachable THEN the system SHALL return network error with retry suggestion
@@ -234,10 +234,10 @@ Complex requirements for unwanted behavior also include the If-Then keywords.
 
 **Acceptance Criteria:**
 
-- WHEN "agentic-knowledge init {docset-id}" executed AND docset has web_sources THEN the system SHALL download all configured sources
-- WHEN "agentic-knowledge refresh" executed without docset-id THEN the system SHALL refresh all docsets with web_sources
-- WHEN "agentic-knowledge refresh {docset-id}" executed THEN the system SHALL refresh only specified docset
-- WHEN "agentic-knowledge status" executed THEN the system SHALL show last update time for each web source
+- WHEN "npx @codemcp/knowledge init {docset-id}" executed AND docset has web_sources THEN the system SHALL download all configured sources
+- WHEN "npx @codemcp/knowledge refresh" executed without docset-id THEN the system SHALL refresh all docsets with web_sources
+- WHEN "npx @codemcp/knowledge refresh {docset-id}" executed THEN the system SHALL refresh only specified docset
+- WHEN "npx @codemcp/knowledge status" executed THEN the system SHALL show last update time for each web source
 - WHEN network is unavailable during refresh THEN the system SHALL report error but continue with other sources
 - The system SHALL show progress indicators for long-running downloads
 

README.md

@@ -40,14 +40,14 @@ Add to your coding agent config something along the lines of
 
 ```bash
 # For a Git repository
-npx agentic-knowledge-mcp create \
+npx @codemcp/knowledge create \
   --preset git-repo \
   --id react-docs \
   --name "React Documentation" \
   --url https://github.com/facebook/react.git
 
 # Initialize (downloads the docs)
-npx agentic-knowledge-mcp init react-docs
+npx @codemcp/knowledge init react-docs
 
 # The MCP server starts automatically when Claude Desktop launches
 ```

TESTING.md

@@ -5,28 +5,32 @@ This guide shows you how to manually test the agentic knowledge MCP server to ve
 ## Quick Test with curl/manual MCP calls
 
 1. **Start the server in a project with configuration**:
+
 ```bash
 cd your-project-with-knowledge-config
-npx agentic-knowledge
+npx @codemcp/knowledge
 ```
 
 2. **Test tools/list request**:
+
 ```bash
-echo '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' | npx agentic-knowledge
+echo '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' | npx @codemcp/knowledge
 ```
 
 Expected response: JSON with `search_docs` and `list_docsets` tools.
 
 3. **Test list_docsets**:
+
 ```bash
-echo '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_docsets","arguments":{}}}' | npx agentic-knowledge
+echo '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_docsets","arguments":{}}}' | npx @codemcp/knowledge
 ```
 
 Expected response: List of available docsets from your configuration.
 
 4. **Test search_docs**:
+
 ```bash
-echo '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"search_docs","arguments":{"docset_id":"your-docset-id","keywords":"search terms","generalized_keywords":"related terms"}}}' | npx agentic-knowledge
+echo '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"search_docs","arguments":{"docset_id":"your-docset-id","keywords":"search terms","generalized_keywords":"related terms"}}}' | npx @codemcp/knowledge
 ```
 
 Expected response: Intelligent search instructions based on your template.
@@ -53,13 +57,15 @@ This ensures LLMs understand how to properly use the search functionality.
 ## Performance Validation
 
 The tests validate:
+
 - ✅ Server creation: <1ms
-- ✅ E2E response time: <1ms (well under 10ms requirement)  
+- ✅ E2E response time: <1ms (well under 10ms requirement)
 - ✅ Memory usage: <1MB
 
 ## What the Tests Cover
 
 ### Unit Tests (84 + 11 = 95)
+
 - Configuration discovery and loading
 - YAML parsing and validation
 - Path calculation and resolution
@@ -68,6 +74,7 @@ The tests validate:
 - Server lifecycle and caching
 
 ### E2E Tests (12)
+
 - MCP protocol compliance
 - Real client-server communication
 - Tool schema validation
@@ -79,13 +86,15 @@ The tests validate:
 ## Expected Test Output
 
 All tests should pass:
+
 ```
 Test Files  1 passed (1)
      Tests  12 passed (12)
   Duration  ~1.5s
 ```
 
 If any tests fail, check:
+
 1. All dependencies installed (`pnpm install`)
 2. Project built successfully (`pnpm build`)
-3. No configuration conflicts in test environment
+3. No configuration conflicts in test environment