Claude Imagine

# Modular Testing Guide ## Running Tests ### Run Single Test: ```bash node test-modular.js 1 # Test 1: Query Session ID node test-modular.js 2 # Test 2: Multiple Instances node test-modular.js 3 # Test 3: Different MCP Config node test-modular.js 4 # Test 4: Different Tools node test-modular.js 5 # Test 5: Different Agents node test-modular.js 6 # Test 6: Project Config node test-modular.js 7 # Test 7: Expert Instance ``` ### Run All Tests: ```bash node test-modular.js ``` ## Test Descriptions ### Test 1: Query Session ID ✅ **Goal:** Extract session ID from Claude output **Command:** `node test-modular.js 1` **Expected:** UUID string in JSON output ### Test 2: Multiple Instances **Goal:** Verify multiple instances can run **Command:** `node test-modular.js 2` **Expected:** Both instances complete with different session IDs ### Test 3: Different MCP Config **Goal:** Each instance has different MCP servers **Command:** `node test-modular.js 3` **Expected:** Instance sees imagine MCP tools ### Test 4: Different Tools **Goal:** Each instance has different tool sets **Command:** `node test-modular.js 4` **Expected:** Instance 1 has Read, Instance 2 has Write ### Test 5: Different Agents **Goal:** Each instance has different agents **Command:** `node test-modular.js 5` **Expected:** Each instance has different agents available ### Test 6: Project Config **Goal:** Project-specific configuration works **Command:** `node test-modular.js 6` **Expected:** Project settings are used ### Test 7: Expert Instance **Goal:** Create specialized expert instance **Command:** `node test-modular.js 7` **Expected:** Expert shows specialized behavior ## Testing Strategy 1. **Run one test at a time** 2. **Verify it passes before moving on** 3. **Fix any issues before proceeding** 4. **Document results** ## Next Steps After Testing 1. Build plugin library 2. Build skills library 3. Create expert instance templates 4. Document best practices