# Workflow Examples
This document provides detailed examples of common workflows using the AI Collaboration MCP Server.
## Complete Development Cycle
### Scenario: Implementing a Lexer for the Compiler
#### 1. CTO Creates the Task
```javascript
// Gemini (CTO)
@ai-collab send_directive {
"taskId": "COMP-001-lexer",
"title": "Implement Lexer for Compiler",
"specification": "Create a lexer that tokenizes source code into a stream of tokens. The lexer should handle all language constructs including keywords, operators, literals, and identifiers.",
"requirements": [
"Support single and multi-line comments",
"Handle string literals with escape sequences",
"Recognize all operators: +, -, *, /, =, ==, !=, <, >, <=, >=",
"Identify keywords: if, else, while, for, function, return, var, const",
"Track line and column numbers for error reporting",
"Implement as a separate module with clean interface"
],
"acceptanceCriteria": [
"Correctly tokenizes all test cases",
"Provides accurate position information",
"Handles edge cases (empty files, long identifiers)",
"Performance: tokenizes 1000 lines in < 100ms",
"100% test coverage for token types"
]
}
```
#### 2. Developer Checks and Starts Work
```javascript
// Claude (Developer)
@ai-collab get_pending_tasks {"role": "developer"}
// Response shows COMP-001-lexer is pending
// Developer reviews the full task:
@ai-collab ai-collab://tasks/COMP-001-lexer
// Developer has a clarification question:
@ai-collab ask_question {
"taskId": "COMP-001-lexer",
"question": "Should the lexer support Unicode identifiers, or ASCII only?",
"context": {
"consideration": "Unicode support affects complexity and performance",
"examples": "Variables like 'пользователь' or '用户'"
}
}
```
#### 3. CTO Answers Question
```javascript
// Gemini (CTO) checks for questions
@ai-collab get_pending_tasks {"role": "cto"}
// Sees question Q-1234567890 and answers:
@ai-collab answer_question {
"questionId": "Q-1234567890",
"answer": "Start with ASCII-only for MVP. We can add Unicode support in a future iteration. Please add a TODO comment where Unicode support would be implemented."
}
```
#### 4. Developer Submits Implementation
```javascript
// Claude (Developer) after implementing
@ai-collab submit_work {
"taskId": "COMP-001-lexer",
"files": {
"src/lexer/lexer.ts": "import { Token, TokenType } from './types';\n\nexport class Lexer {\n // Implementation...",
"src/lexer/types.ts": "export enum TokenType {\n // Token type definitions...",
"src/lexer/index.ts": "export { Lexer } from './lexer';\nexport * from './types';",
"tests/lexer.test.ts": "import { Lexer } from '../src/lexer';\n\ndescribe('Lexer', () => {\n // Test cases..."
},
"summary": "Implemented lexer with all required features. ASCII-only as discussed, with TODO markers for future Unicode support. All tests passing with 100% coverage.",
"testResults": {
"passed": 47,
"failed": 0,
"coverage": {
"statements": 100,
"branches": 100,
"functions": 100,
"lines": 100
}
}
}
```
#### 5. CTO Reviews Submission
```javascript
// Gemini (CTO)
@ai-collab get_pending_tasks {"role": "cto"}
// Reviews the submission
@ai-collab ai-collab://tasks/COMP-001-lexer
// Submits review with minor feedback
@ai-collab submit_review {
"taskId": "COMP-001-lexer",
"status": "needs_revision",
"feedback": "Excellent implementation overall. Clean architecture and comprehensive tests. A few minor items need attention before approval.",
"actionItems": [
"Add handling for hexadecimal number literals (0x1234)",
"The error message for unterminated strings should include the starting line number",
"Consider extracting magic numbers (like max identifier length) to constants"
]
}
```
#### 6. Developer Makes Revisions
```javascript
// Claude (Developer) sees revision request
@ai-collab get_pending_tasks {"role": "developer"}
// Makes the requested changes and resubmits
@ai-collab submit_work {
"taskId": "COMP-001-lexer",
"files": {
"src/lexer/constants.ts": "export const MAX_IDENTIFIER_LENGTH = 255;\nexport const HEX_PREFIX = '0x';",
"src/lexer/lexer.ts": "// Updated implementation with hex support and constants",
"tests/lexer.test.ts": "// Added tests for hex literals"
},
"summary": "Addressed all feedback: Added hex literal support, improved error messages with line numbers, extracted constants. Added 5 new test cases.",
"testResults": {
"passed": 52,
"failed": 0,
"coverage": {
"statements": 100,
"branches": 100,
"functions": 100,
"lines": 100
}
}
}
```
#### 7. CTO Approves
```javascript
// Gemini (CTO)
@ai-collab submit_review {
"taskId": "COMP-001-lexer",
"status": "approved",
"feedback": "Perfect! All issues addressed. The lexer is well-structured and thoroughly tested. Ready for integration.",
"actionItems": []
}
```
## Parallel Task Management
### Working on Multiple Features
```javascript
// CTO creates multiple tasks at once
const tasks = [
{
taskId: "COMP-002-parser",
title: "Implement Recursive Descent Parser",
dependencies: ["COMP-001-lexer"]
},
{
taskId: "COMP-003-ast",
title: "Design AST Node Structure",
dependencies: []
},
{
taskId: "FEAT-001-repl",
title: "Create REPL for Testing",
dependencies: ["COMP-001-lexer"]
}
];
// Create all tasks
for (const task of tasks) {
@ai-collab send_directive { ...task }
}
// Developer can work on independent tasks
@ai-collab get_pending_tasks {"role": "developer"}
// Shows COMP-003-ast (no dependencies) and others
```
## Project State Management
### Establishing Architecture Decisions
```javascript
// CTO updates project architecture
@ai-collab update_project_state {
"component": "architecture",
"state": {
"decisions": [
{
"id": "ADR-001",
"title": "Use Recursive Descent Parser",
"rationale": "Simpler to implement and debug than table-driven approach",
"consequences": "May be slower for very large files"
}
],
"patterns": [
"Visitor pattern for AST traversal",
"Builder pattern for IR construction"
]
}
}
// Developer can reference these decisions
@ai-collab ai-collab://project-state/architecture
```
### Code Standards
```javascript
// CTO establishes coding standards
@ai-collab update_project_state {
"component": "standards",
"state": {
"typescript": {
"strictMode": true,
"noImplicitAny": true,
"naming": {
"interfaces": "PascalCase with 'I' prefix",
"types": "PascalCase",
"enums": "PascalCase"
}
},
"testing": {
"framework": "Jest",
"structure": "Colocated with source files",
"naming": "*.test.ts",
"coverage": {
"statements": 90,
"branches": 85,
"functions": 90,
"lines": 90
}
}
}
}
```
## Monitoring Progress
### Checking Project Status
```javascript
// View all tasks and their states
@ai-collab ai-collab://tasks
// View specific component progress
@ai-collab ai-collab://project-state/lexer
// Check recent communications
@ai-collab ai-collab://logs
```
### Generating Reports
```javascript
// CTO can track velocity and progress
const recentLogs = @ai-collab ai-collab://logs
// Analyze:
// - Tasks completed this week
// - Average review cycles per task
// - Common revision reasons
// - Question frequency
```
## Error Handling Scenarios
### Build Failures
```javascript
// Developer encounters build issue
@ai-collab submit_work {
"taskId": "COMP-004-optimizer",
"files": { /* partial implementation */ },
"summary": "Partial implementation - encountering circular dependency issue with current module structure",
"testResults": {
"buildFailed": true,
"error": "Cannot resolve circular dependency between optimizer and analyzer"
}
}
// Include question about architectural change
@ai-collab ask_question {
"taskId": "COMP-004-optimizer",
"question": "Should we refactor to extract shared interfaces to a separate module?",
"context": {
"issue": "Circular dependency between optimizer and analyzer",
"proposedSolution": "Create 'interfaces' module"
}
}
```
## Best Practices Demonstrated
1. **Clear Task Definitions**: Include specific requirements and measurable acceptance criteria
2. **Iterative Communication**: Use questions for clarification before implementation
3. **Comprehensive Submissions**: Include all relevant files and test results
4. **Actionable Feedback**: Specific items to address in reviews
5. **State Documentation**: Keep architectural decisions and standards in project state
6. **Progress Tracking**: Regular status checks and log reviews
