/f5-agent - Unified AI Agent Command

Consolidated command that replaces:

/f5-persona → /f5-agent persona
/f5-suggest → /f5-agent suggest

ARGUMENTS

$ARGUMENTS

MODE DETECTION

Pattern	Mode	Description
`persona <action>`	PERSONA	Persona management
`suggest [topic]`	SUGGEST	Smart suggestions
`<other>` (default)	AGENT	Agent invocation

MODE: PERSONA (from /f5-persona)

`/f5-agent persona <action>`

Manage and activate specialized personas.

Action	Description
`list`	List all available personas
`show <name>`	Show persona details
`activate <name>`	Activate a persona
`deactivate`	Return to default
`auto [on\|off]`	Toggle auto-activation
`chain <task>`	Chain personas for task
`status`	Show current status

Available Personas:

Persona	Emoji	Focus
architect	🏗️	System design
backend	⚙️	APIs, services
frontend	🎨	UI/UX, components
database	🗄️	Schema, queries
security	🔒	OWASP, auth
qa	🧪	Testing, coverage
performance	⚡	Optimization
reviewer	👀	Code review
devops	🚀	CI/CD, infra
analyst	📊	Requirements
developer	💻	General dev (default)

Examples:

/f5-agent persona list
/f5-agent persona activate security
/f5-agent persona chain "build auth system"

MODE: SUGGEST (from /f5-suggest)

`/f5-agent suggest [topic]`

Get context-aware command suggestions.

Action	Description
(default)	Suggestions for current phase
`<topic>`	Suggestions for specific topic
`next`	Suggest next action
`chains`	Show command chains

Options:

--all - Show all command levels
--v{1-5} - Verbosity level

Examples:

/f5-agent suggest
/f5-agent suggest api
/f5-agent suggest next
/f5-agent suggest chains

MODE: AGENT (Default)

Invoke specialized agents for specific tasks.

OVERVIEW

SPECIALIZED AGENTS
═══════════════════════════════════════════════════════════════════

  Agents are task-specific AI assistants:

  ⚡ code_generator     │ 🧪 test_writer      │ 👀 code_reviewer
  🔧 refactorer         │ 📝 documenter       │ 🐛 debugger
  🔄 migrator           │ 🛡️ security_scanner │ 📊 performance_analyzer
  🌐 api_designer       │ 🏗️ system_architect

═══════════════════════════════════════════════════════════════════

SUBCOMMANDS

Command	Description
`list`	List all agents
`invoke <agent> [input]`	Invoke an agent
`pipeline <name>`	Run agent pipeline
`status`	Show active agents
`help <agent>`	Show agent details

QUICK INVOCATION

Instead of using the full command, you can invoke agents directly using the @f5: pattern:

Direct Pattern

@f5:security "review authentication flow"
@f5:architect "design payment system"
@f5:qa "generate tests for PaymentService"
@f5:debug "fix TypeError in OrderController"

Multi-Agent

@f5:security @f5:architect "review and redesign auth system"

With Options

@f5:qa --coverage 90 "test all services"
@f5:security --deep "full security audit"

Available Aliases

Short	Full	Agent
@f5:sec	@f5:security	security_scanner
@f5:arch	@f5:architect	system_architect
@f5:be	@f5:backend	code_generator
@f5:fe	@f5:frontend	code_generator
@f5:qa	@f5:test	test_writer
@f5:bug	@f5:debug	debugger
@f5:doc	@f5:docs	documenter
@f5:perf	@f5:performance	performance_analyzer
@f5:ref	@f5:refactor	refactorer
@f5:api	-	api_designer

Supported Options

Option	Description
`--deep`	Thorough analysis
`--quick`	Fast check
`--coverage <n>`	Target coverage percentage
`--fix`	Auto-fix issues if possible
`--verbose`	Detailed output

ACTION: LIST_AGENTS

`/f5-agent list`

## 🤖 Available Agents

### Code Agents
| Agent | Emoji | Purpose |
|-------|-------|---------|
| code_generator | ⚡ | Generate production code |
| refactorer | 🔧 | Improve code quality |
| migrator | 🔄 | Handle migrations |

### Quality Agents
| Agent | Emoji | Purpose |
|-------|-------|---------|
| test_writer | 🧪 | Generate test suites |
| code_reviewer | 👀 | Review code quality |
| security_scanner | 🛡️ | Find vulnerabilities |
| performance_analyzer | 📊 | Analyze performance |

### Support Agents
| Agent | Emoji | Purpose |
|-------|-------|---------|
| debugger | 🐛 | Diagnose and fix bugs |
| documenter | 📝 | Generate documentation |
| api_designer | 🌐 | Design APIs |

### Quick Invoke
```bash
/f5-agent invoke code_generator "user registration service"


---

## ACTION: INVOKE_AGENT

### `/f5-agent invoke <agent> [input]`

Invoke a specific agent with input:

```markdown
## ⚡ Agent Invoked: code_generator

### Input
> "user registration service"

### Agent Processing

Analyzing requirements...
Loading project patterns...
Generating code...


### Output

```typescript
// src/user/user-registration.service.ts
import { Injectable, ConflictException } from '@nestjs/common';
import { InjectRepository } from '@nestjs/typeorm';
import { Repository } from 'typeorm';
import { User } from './entities/user.entity';
import { RegisterDto } from './dto/register.dto';
import * as bcrypt from 'bcrypt';

@Injectable()
export class UserRegistrationService {
  constructor(
    @InjectRepository(User)
    private readonly userRepo: Repository<User>,
  ) {}

  async register(dto: RegisterDto): Promise<User> {
    // Check duplicate
    const existing = await this.userRepo.findOne({
      where: { email: dto.email }
    });
    if (existing) {
      throw new ConflictException('Email already registered');
    }

    // Hash password
    const passwordHash = await bcrypt.hash(dto.password, 10);

    // Create user
    const user = this.userRepo.create({
      ...dto,
      password: passwordHash,
    });

    return this.userRepo.save(user);
  }
}

Agent Notes

Included duplicate email check
Password hashing with bcrypt
Follows NestJS patterns
Includes proper error handling

Follow-up Actions

/f5-agent invoke test_writer "for UserRegistrationService"
/f5-agent invoke documenter "for UserRegistrationService"


---

## ACTION: RUN_PIPELINE

### `/f5-agent pipeline <name>`

Run a predefined agent pipeline:

```markdown
## 🔗 Running Pipeline: feature_development

### Pipeline Steps

| Step | Agent | Status |
|------|-------|--------|
| 1 | 🌐 api_designer | ⏳ Running |
| 2 | ⚡ code_generator | ⏸️ Waiting |
| 3 | 🧪 test_writer | ⏸️ Waiting |
| 4 | 👀 code_reviewer | ⏸️ Waiting |
| 5 | 📝 documenter | ⏸️ Waiting |

### Step 1: API Designer

Designing API endpoints...

```yaml
# Generated API Design
paths:
  /users/register:
    post:
      summary: Register new user
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/RegisterDto'
      responses:
        201:
          description: User created
        409:
          description: Email already exists

✅ Step 1 Complete

Step 2: Code Generator

Generating implementation...

[continues for each step]


---

## AVAILABLE PIPELINES

### `/f5-agent pipeline list`

| Pipeline | Description | Agents |
|----------|-------------|--------|
| `feature_development` | Full feature development | api_designer → code_generator → test_writer → code_reviewer → documenter |
| `bug_fix` | Bug fixing | debugger → code_generator → test_writer → code_reviewer |
| `code_quality` | Code quality improvement | code_reviewer → refactorer → security_scanner → performance_analyzer |
| `security_audit` | Security audit | security_scanner → code_reviewer → code_generator → test_writer |
| `api_design` | API design and docs | api_designer → documenter → code_reviewer |
| `refactoring` | Code refactoring | code_reviewer → refactorer → test_writer → documenter |
| `migration` | Migration | migrator → test_writer → code_reviewer → documenter |

---

## ACTION: SHOW_STATUS

### `/f5-agent status`

```markdown
## 📊 Agent Status

### Active Agents
| Agent | Task | Progress |
|-------|------|----------|
| code_generator | Generate payment service | 75% |

### Recent Activity
| Time | Agent | Task | Result |
|------|-------|------|--------|
| 5m ago | 🧪 test_writer | Generate tests for UserService | ✅ Complete |
| 15m ago | ⚡ code_generator | Generate UserService | ✅ Complete |
| 30m ago | 🌐 api_designer | Design User API | ✅ Complete |

### Running Pipelines
| Pipeline | Status | Current Step |
|----------|--------|--------------|
| feature_development | 🔄 Running | 4/5 (code_reviewer) |

AGENT DETAILS

⚡ code_generator

Purpose: Generates production-ready code from specifications

Capabilities:

Generate code from description
Follow project patterns
Apply best practices
Include error handling
Type-safe implementations

Usage:

/f5-agent invoke code_generator "payment service"
/f5-agent invoke code_generator --spec ./specs/order.yaml

🧪 test_writer

Purpose: Generates comprehensive test suites

Capabilities:

Generate unit tests
Generate integration tests
Identify edge cases
Create mocks and fixtures
Achieve coverage targets (80%)

Usage:

/f5-agent invoke test_writer "for payment.service.ts"
/f5-agent invoke test_writer --coverage 90

👀 code_reviewer

Purpose: Reviews code for quality, security, and best practices

Capabilities:

Identify code issues
Security vulnerability detection
Performance analysis
Best practice enforcement
Constructive feedback

Usage:

/f5-agent invoke code_reviewer src/payment/
/f5-agent invoke code_reviewer --pr 123

🔧 refactorer

Purpose: Improves code quality through refactoring

Capabilities:

Identify code smells
Apply design patterns
Reduce complexity
Extract methods/classes
Improve naming

Usage:

/f5-agent invoke refactorer src/services/
/f5-agent invoke refactorer "OrderService class"

📝 documenter

Purpose: Generates and maintains documentation

Capabilities:

Generate API documentation
Create README files
Write inline comments
Generate JSDoc/TSDoc
Create architecture docs

Usage:

/f5-agent invoke documenter "for UserService"
/f5-agent invoke documenter --type api

🐛 debugger

Purpose: Diagnoses and fixes bugs systematically

Capabilities:

Root cause analysis
Error interpretation
Fix suggestion
Regression prevention
Debug logging

Methodology:

Reproduce the issue
Isolate the problem
Identify root cause
Implement fix
Verify fix
Add regression test

Usage:

/f5-agent invoke debugger "TypeError: Cannot read property 'id' of undefined"
/f5-agent invoke debugger --stack-trace ./error.log

🔄 migrator

Purpose: Handles code and data migrations

Capabilities:

Version upgrade assistance
Breaking change handling
Data migration scripts
Dependency updates
API migration

Usage:

/f5-agent invoke migrator "upgrade React 17 to 18"
/f5-agent invoke migrator --from v1 --to v2

🛡️ security_scanner

Purpose: Scans for security vulnerabilities

Capabilities:

OWASP Top 10 check
Dependency vulnerability scan
Code security analysis
Secret detection
Security recommendations

Checks:

SQL Injection
XSS
CSRF
Authentication issues
Sensitive data exposure
Dependency vulnerabilities

Usage:

/f5-agent invoke security_scanner src/auth/
/f5-agent invoke security_scanner --full

📊 performance_analyzer

Purpose: Analyzes and optimizes performance

Capabilities:

Identify bottlenecks
Complexity analysis (Big O)
Query optimization
Caching recommendations
Memory analysis

Analysis Types:

Time complexity
Space complexity
N+1 queries
Memory leaks
Slow queries

Usage:

/f5-agent invoke performance_analyzer src/services/
/f5-agent invoke performance_analyzer --profile ./metrics.json

🌐 api_designer

Purpose: Designs RESTful and GraphQL APIs

Capabilities:

RESTful API design
GraphQL schema design
OpenAPI specification
Versioning strategy
Error handling design

Principles:

Consistent naming
Proper HTTP methods
Meaningful status codes
Pagination support
Filtering/sorting

Usage:

/f5-agent invoke api_designer "user management endpoints"
/f5-agent invoke api_designer --graphql "order schema"

🏗️ system_architect

Purpose: High-level system design, architecture patterns, scalability

Capabilities:

Design system architecture
Evaluate architectural trade-offs
Recommend design patterns
Plan for scalability
Create architecture diagrams
Document architecture decisions

Core Mindset:

Primary Question: "How will this scale and evolve?"
Core Belief: Systems must be designed for change
Approach: Think in diagrams, patterns, and trade-offs

Usage:

/f5-agent invoke system_architect "design user authentication system"
/f5-agent invoke system_architect "evaluate microservices vs monolith"
@f5:architect "design payment system"
@f5:arch --diagram "payment processing flow"

EXAMPLES

# List agents
/f5-agent list

# Generate code
/f5-agent invoke code_generator "payment service"

# Write tests
/f5-agent invoke test_writer "for payment.service.ts"

# Review code
/f5-agent invoke code_reviewer src/payment/

# Run full pipeline
/f5-agent pipeline feature_development

# Security scan
/f5-agent invoke security_scanner src/auth/

# Debug error
/f5-agent invoke debugger "TypeError: Cannot read property 'id' of undefined"

# Run bug fix pipeline
/f5-agent pipeline bug_fix

# Performance analysis
/f5-agent invoke performance_analyzer --profile ./metrics.json

AUTO-SELECTION RULES

Agents can be auto-selected based on task keywords:

Keywords	Agent
generate, implement, create, write code	code_generator
test, unit test, integration test	test_writer
review, check code, pr review	code_reviewer
refactor, improve, clean up	refactorer
document, readme, api docs	documenter
debug, fix bug, error, troubleshoot	debugger
migrate, upgrade, update version	migrator
security, vulnerability, audit	security_scanner
performance, optimize, slow	performance_analyzer
api, endpoint, rest, graphql	api_designer
architecture, system design, scalability, trade-off	system_architect

AGENTS VS PERSONAS

Aspect	Agents	Personas
Focus	Task-specific	Role-based
Question	"What do I do?"	"Who am I?"
Scope	Single task	Entire session
Output	Specific artifact	General guidance
Example	test_writer	qa engineer

Use agents for specific tasks, personas for overall mindset.

GATE-AGENT RECOMMENDATIONS

Gate	Recommended Agents
D2-D4	api_designer, documenter
G2	code_generator, test_writer, code_reviewer
G3	test_writer, security_scanner, performance_analyzer
G4	security_scanner, documenter

/f5-agent

/f5-agent - Unified AI Agent Command

ARGUMENTS

MODE DETECTION

MODE: PERSONA (from /f5-persona)

/f5-agent persona <action>

MODE: SUGGEST (from /f5-suggest)

/f5-agent suggest [topic]

MODE: AGENT (Default)

OVERVIEW

SUBCOMMANDS

QUICK INVOCATION

Direct Pattern

Multi-Agent

With Options

Available Aliases

Supported Options

ACTION: LIST_AGENTS

/f5-agent list

Agent Notes

Follow-up Actions

Step 2: Code Generator

AGENT DETAILS

⚡ code_generator

🧪 test_writer

👀 code_reviewer

🔧 refactorer

📝 documenter

🐛 debugger

🔄 migrator

🛡️ security_scanner

📊 performance_analyzer

🌐 api_designer

🏗️ system_architect

EXAMPLES

AUTO-SELECTION RULES

AGENTS VS PERSONAS

GATE-AGENT RECOMMENDATIONS

SEE ALSO

`/f5-agent persona <action>`

`/f5-agent suggest [topic]`

`/f5-agent list`