Initial commit

commands/start.md

@@ -0,0 +1,94 @@
+---
+description: Start intelligent coding workflow with automated testing and code review
+argument-hint: <your coding requirement>
+model: inherit
+---
+
+# Droids Coding Workflow
+
+## User Requirement
+
+$ARGUMENTS
+
+## Workflow Overview
+
+You will coordinate specialized agents to complete this coding task with quality assurance. After implementing the core functionality yourself, you MUST use the following agents:
+
+1. **test-engineer** agent - For comprehensive testing
+2. **code-reviewer** agent - For quality and security review
+3. **doc-writer** agent (optional) - For documentation
+
+## Step-by-Step Process
+
+### Step 1: Analyze & Plan (You do this)
+
+1. Read CLAUDE.md if it exists to understand project standards
+2. Analyze the codebase structure relevant to the requirement
+3. Create a clear implementation plan
+4. Use TodoWrite to track your plan
+
+### Step 2: Implement Core Functionality (You do this)
+
+1. Implement the required functionality directly
+2. Make necessary file changes
+3. Follow coding standards from CLAUDE.md
+4. Handle edge cases and errors properly
+
+### Step 3: Use test-engineer Agent (REQUIRED)
+
+After implementation, use the **test-engineer** agent to write and run comprehensive tests.
+
+Provide the agent with:
+- The implemented code/feature
+- Test requirements: backend tests (unit, integration, API), frontend tests (component, E2E), contract tests
+
+The agent should:
+- Write tests following project patterns
+- Run all tests
+- Report coverage and results
+
+### Step 4: Use code-reviewer Agent (REQUIRED)
+
+After tests pass, use the **code-reviewer** agent to assess code quality.
+
+Provide the agent with:
+- The implemented code
+- Test results from step 3
+
+The agent should check:
+- Security vulnerabilities
+- Code quality and maintainability
+- Performance issues
+- Standards compliance with CLAUDE.md
+
+### Step 5: Iterate if Needed
+
+- If tests fail: Fix the code and re-run test-engineer agent
+- If review finds critical issues: Address them, then re-test and re-review
+- Maximum 3 iterations
+
+### Step 6: Use doc-writer Agent (Optional)
+
+If documentation is needed, use the **doc-writer** agent.
+
+The agent should:
+- Add inline comments (JSDoc/docstring) for key functions
+- Create/update API documentation
+- Update README if needed
+
+### Step 7: Report Completion
+
+Provide a summary to the user including:
+- What was implemented
+- Test results (all passing)
+- Code review results (approved)
+- Any documentation created
+
+## Important Notes
+
+- **You must use the agents** - Don't skip the test-engineer and code-reviewer steps
+- Detect user's language and respond in the same language (Chinese/中文 or English)
+- Keep the user informed with clear progress updates
+- Follow CLAUDE.md standards throughout
+
+Start now by analyzing the requirement and creating your implementation plan!