Agent Management

Complete guide to managing AI agents in NikCLI

Overview

NikCLI’s agent management system provides sophisticated control over AI agents, enabling you to orchestrate multiple specialized agents, create custom agents, and manage complex multi-agent workflows. The system supports both individual agent operations and coordinated multi-agent collaboration.

Agent Orchestration

Coordinate multiple agents for complex tasks

Custom Agents

Create and deploy specialized agents

Performance Monitoring

Track agent performance and effectiveness

Workflow Management

Design and execute multi-agent workflows

Core Agent Commands

/agents - Agent Discovery and Status

Display comprehensive information about available agents and their current status.

Basic Agent Listing

# List all available agents
/agents

# Show detailed agent information
/agents --detailed
/agents --verbose

# Filter agents by capability
/agents --frontend
/agents --backend
/agents --devops

Output Example:

Available Agents:
┌─────────────────┬──────────────┬─────────┬─────────────┐
│ Agent           │ Specialization│ Status  │ Performance │
├─────────────────┼──────────────┼─────────┼─────────────┤
│ universal-agent │ Full-stack   │ Active  │ 95.2%       │
│ react-expert    │ React/UI     │ Idle    │ 92.8%       │
│ backend-agent   │ APIs/Services│ Active  │ 89.5%       │
│ devops-agent    │ CI/CD/Deploy │ Idle    │ 87.3%       │
└─────────────────┴──────────────┴─────────┴─────────────┘

/agent - Individual Agent Execution

Execute tasks with specific agents, providing fine-grained control over agent selection and behavior.

Direct Agent Invocation

# Execute with universal agent
/agent universal-agent "create a React login component"

# Execute with specialized agent
/agent react-expert "optimize this component for performance"
/agent backend-agent "create REST API for user management"
/agent devops-agent "set up CI/CD pipeline for deployment"

Agent Selection Guide:

• universal-agent: Best for general tasks requiring multiple skills
• react-expert: Optimal for React/Next.js specific tasks
• backend-agent: Ideal for API, database, and server-side logic
• devops-agent: Perfect for deployment, monitoring, and infrastructure

Agent Configuration

# Configure agent parameters
/agent react-expert --temperature 0.3 "refactor component"
/agent backend-agent --max-tokens 4000 "design database schema"

# Set agent context
/agent universal-agent --context "src/components/" "analyze code"
/agent devops-agent --workspace "infrastructure/" "update configs"

# Agent execution modes
/agent react-expert --mode creative "design new UI pattern"
/agent backend-agent --mode conservative "fix security issue"

Agent Chaining

# Sequential agent execution
/agent backend-agent "create API endpoint" | 
/agent react-expert "create UI component for API"

# Conditional agent chaining
/agent universal-agent "analyze project structure" --then-if-frontend 
/agent react-expert "optimize components"

# Agent handoff with context
/agent backend-agent "implement authentication" --handoff-to 
/agent react-expert --context "auth implementation"

/auto - Autonomous Multi-Agent Execution

Enable autonomous mode where agents work together with minimal human intervention.

Basic Autonomous Mode

# Simple autonomous task
/auto "create a todo application with React and Node.js"

# Complex autonomous project
/auto "build a e-commerce platform with authentication, 
       payment processing, and admin dashboard"

# Autonomous bug fixing
/auto "identify and fix all TypeScript errors in the project"

/parallel - Parallel Agent Execution

Execute multiple agents simultaneously on related or independent tasks.

Parallel Task Distribution

# Parallel execution with multiple agents
/parallel "react-expert,backend-agent" 
          "create user dashboard with API integration"

# Distribute tasks across agents
/parallel --distribute "optimize frontend performance" 
          --agents "react-expert,universal-agent"

# Parallel with different tasks
/parallel --tasks "frontend:react-expert:create UI components"
                 "backend:backend-agent:implement API endpoints"
                 "devops:devops-agent:setup deployment pipeline"

Parallel Coordination

# Coordinated parallel execution
/parallel --coordinated 
          --agents "react-expert,backend-agent"
          "build authentication system"

# Parallel with synchronization points
/parallel --sync-points "design,implementation,testing"
          --agents "universal-agent,react-expert"
          "create component library"

# Parallel with conflict resolution
/parallel --resolve-conflicts auto
          --agents "react-expert,universal-agent"
          "refactor application architecture"

Parallel Monitoring

# Monitor parallel execution
/parallel status
/parallel --progress --per-agent

# Parallel execution control
/parallel pause react-expert
/parallel resume backend-agent
/parallel rebalance

# Parallel results aggregation
/parallel results
/parallel --merge-outputs --summary

Advanced Agent Management

/factory - Agent Factory Dashboard

Access the agent factory for creating, configuring, and managing custom agents.

Agent Creation Interface

# Open agent factory dashboard
/factory

# Create new agent interactively
/factory create
/factory --wizard --guided-creation

# Agent creation from template
/factory template
/factory --template react-specialist
/factory --template api-developer

/create-agent - Custom Agent Creation

Create specialized agents tailored to specific domains or use cases.

Domain-Specific Agents

# Create React testing specialist
/create-agent "React Testing Expert" 
              --domain "frontend-testing"
              --skills "jest,react-testing-library,cypress"
              --context "testing best practices"

# Create API security specialist
/create-agent "API Security Analyst"
              --domain "backend-security"
              --skills "owasp,jwt,oauth,encryption"
              --tools "security-scanner,vulnerability-assessment"

# Create DevOps automation specialist
/create-agent "Infrastructure Automation Expert"
              --domain "devops-automation"
              --skills "terraform,kubernetes,ansible,monitoring"
              --cloud-providers "aws,gcp,azure"

Task-Specific Agents

# Create code review agent
/create-agent "Code Review Specialist"
              --task "code-review"
              --focus "security,performance,maintainability"
              --languages "typescript,python,go"

# Create documentation agent
/create-agent "Documentation Expert"
              --task "documentation"
              --formats "markdown,jsdoc,api-docs"
              --tools "mintlify,gitbook,confluence"

# Create performance optimization agent
/create-agent "Performance Optimizer"
              --task "optimization"
              --targets "bundle-size,runtime,memory"
              --frameworks "react,vue,angular"

Industry-Specific Agents

# Create fintech compliance agent
/create-agent "Fintech Compliance Expert"
              --industry "fintech"
              --compliance "pci-dss,gdpr,sox"
              --focus "payment-processing,audit-trails"

# Create healthcare data agent
/create-agent "Healthcare Data Specialist"
              --industry "healthcare"
              --compliance "hipaa,fhir,hl7"
              --focus "patient-data,privacy,interoperability"

# Create e-commerce optimization agent
/create-agent "E-commerce Optimizer"
              --industry "e-commerce"
              --focus "conversion,ux,performance,seo"
              --platforms "shopify,woocommerce,magento"

/launch-agent - Blueprint-Based Agent Deployment

Deploy agents from pre-configured blueprints or saved configurations.

Blueprint Deployment

# Launch from saved blueprint
/launch-agent blueprint-id "implement user authentication"

# Launch with configuration override
/launch-agent react-testing-blueprint 
              --override temperature=0.2
              "create comprehensive test suite"

# Launch multiple agents from blueprint
/launch-agent full-stack-blueprint
              --agents "frontend,backend,devops"
              "build complete application"

Agent Collaboration and Orchestration

/orchestrate - Multi-Agent Orchestration

Design and execute complex multi-agent workflows with sophisticated coordination.

Workflow Definition

# Define multi-agent workflow
/orchestrate workflow "full-stack-development"
            --phases "design,backend,frontend,testing,deployment"
            --agents "universal-agent,backend-agent,react-expert,devops-agent"

# Create conditional workflow
/orchestrate workflow "adaptive-development"
            --condition "if typescript then add type-checking phase"
            --condition "if react then optimize for performance"

# Workflow with dependencies
/orchestrate workflow "microservices-deployment"
            --dependencies "database:before:api"
            --dependencies "api:before:frontend"
            --dependencies "frontend:before:e2e-tests"

Agent Coordination

# Coordinate agent handoffs
/orchestrate handoff "backend-agent" "react-expert"
            --context "API implementation completed"
            --next-task "create frontend integration"

# Synchronize agent checkpoints
/orchestrate sync --checkpoint "design-review"
            --agents "react-expert,backend-agent"
            --approval-required

# Agent conflict resolution
/orchestrate resolve-conflict
            --agents "react-expert,universal-agent"
            --conflict-type "implementation-approach"
            --resolution-strategy "expert-preference"

Orchestration Monitoring

# Monitor workflow execution
/orchestrate status
/orchestrate --timeline --progress --bottlenecks

# Workflow performance analysis
/orchestrate analytics
/orchestrate --efficiency --resource-usage --success-rate

# Workflow optimization
/orchestrate optimize
/orchestrate --rebalance --parallel-opportunities --reduce-overhead

/context - Agent Context Management

Manage workspace context and information sharing between agents.

Context Configuration

# Set workspace context
/context workspace src/
/context --paths "src/components/,src/hooks/,src/utils/"

# Set context with filters
/context --include "*.ts,*.tsx"
          --exclude "*.test.*,*.spec.*"
          --max-files 100

# Dynamic context based on task
/context auto --task "react component optimization"
         --smart-selection --relevant-files-only

/stream - Agent Activity Streaming

Monitor and control live agent activity streams for real-time visibility.

Stream Dashboard

# Open stream dashboard
/stream

# View specific agent stream
/stream agent react-expert
/stream --agent backend-agent --verbose

# Multiple agent streams
/stream multi --agents "react-expert,backend-agent"
        --layout split --real-time

Stream Control

# Pause agent stream
/stream pause react-expert

# Resume agent stream
/stream resume backend-agent

# Clear completed streams
/stream clear
/stream --cleanup --completed --older-than 24h

# Stream recording
/stream record --agent universal-agent
        --output "agent-session-$(date).log"

Stream Analytics

# Stream performance metrics
/stream metrics
/stream --throughput --response-time --error-rate

# Stream pattern analysis
/stream patterns
/stream --common-tasks --efficiency-trends --bottlenecks

# Stream alerting
/stream alerts --setup
        --threshold response-time=5s
        --threshold error-rate=10%
        --notify-on failure

Agent Performance and Optimization

/agent-performance - Performance Monitoring

Monitor and analyze agent performance metrics for optimization opportunities.

Performance Metrics

# Overall performance dashboard
/agent-performance dashboard
/agent-performance --metrics --charts --trends

# Agent-specific performance
/agent-performance react-expert
/agent-performance --agent backend-agent --detailed

# Comparative performance analysis
/agent-performance compare
/agent-performance --agents "react-expert,universal-agent"
                   --metric efficiency --time-range 7d

/optimize-agents - Agent Optimization

Optimize agent configuration and performance based on usage patterns and feedback.

Automatic Optimization

# Auto-optimize all agents
/optimize-agents auto
/optimize-agents --auto-tune --performance-based --usage-patterns

# Optimize specific agent
/optimize-agents react-expert
/optimize-agents --agent backend-agent --focus response-time

# Optimization recommendations
/optimize-agents recommend
/optimize-agents --suggestions --impact-analysis --implementation-guide

Configuration Optimization

# Temperature optimization
/optimize-agents temperature
/optimize-agents --temperature-tuning --task-specific

# Context window optimization
/optimize-agents context
/optimize-agents --context-sizing --relevance-filtering

# Tool usage optimization
/optimize-agents tools
/optimize-agents --tool-selection --access-patterns

Learning and Adaptation

# Agent learning from feedback
/optimize-agents learn
/optimize-agents --feedback-integration --pattern-recognition

# Adaptive agent behavior
/optimize-agents adapt
/optimize-agents --behavioral-adaptation --context-awareness

# Continuous improvement
/optimize-agents improve
/optimize-agents --continuous-learning --performance-monitoring

Agent Security and Governance

/agent-security - Security Management

Implement comprehensive security controls and governance for agent operations.

Access Control

# Agent permission management
/agent-security permissions
/agent-security --permissions --agent react-expert --tools --resources

# Role-based access control
/agent-security rbac
/agent-security --roles --assign "developer:react-expert,backend-agent"

# Security policy enforcement
/agent-security policy
/agent-security --policy-check --violations --enforcement

/agent-governance - Governance Framework

Establish governance frameworks for agent development, deployment, and operations.

Governance Policies

# Define governance policies
/agent-governance policy-define
/agent-governance --policy "agent-creation-approval"
                  --policy "performance-standards"
                  --policy "security-requirements"

# Policy enforcement
/agent-governance enforce
/agent-governance --policy-enforcement --violations --remediation

# Policy compliance reporting
/agent-governance compliance
/agent-governance --compliance-report --exceptions --improvements

Quality Standards

# Define quality standards
/agent-governance quality-standards
/agent-governance --code-quality --performance --reliability

# Quality assessment
/agent-governance assess
/agent-governance --agent react-expert --quality-metrics --benchmarks

# Quality improvement plans
/agent-governance improve
/agent-governance --improvement-plans --training --optimization

Risk Management

# Risk assessment
/agent-governance risk-assessment
/agent-governance --identify-risks --impact-analysis --mitigation

# Risk monitoring
/agent-governance risk-monitor
/agent-governance --continuous-monitoring --alerts --dashboard

# Risk mitigation
/agent-governance risk-mitigate
/agent-governance --mitigation-strategies --controls --validation

Advanced Agent Features

/agent-learning - Adaptive Learning System

Enable agents to learn and improve from interactions and feedback.

Learning Configuration

# Enable agent learning
/agent-learning enable react-expert
/agent-learning --learning-mode adaptive --feedback-integration

# Learning data management
/agent-learning data
/agent-learning --training-data --examples --quality-control

# Learning model updates
/agent-learning update
/agent-learning --model-updates --incremental --validation

/agent-collaboration - Inter-Agent Collaboration

Enable sophisticated collaboration patterns between multiple agents.

Collaboration Patterns

# Pair programming mode
/agent-collaboration pair
/agent-collaboration --agents "react-expert,backend-agent"
                     --mode pair-programming
                     --task "implement authentication"

# Code review collaboration
/agent-collaboration review
/agent-collaboration --reviewer "security-agent"
                     --reviewee "backend-agent"
                     --focus "security,performance"

# Mentoring collaboration
/agent-collaboration mentor
/agent-collaboration --mentor "senior-agent"
                     --mentee "junior-agent"
                     --learning-objectives "best-practices"

Knowledge Sharing

# Share agent insights
/agent-collaboration share-insights
/agent-collaboration --from react-expert --to universal-agent
                     --insights "performance-patterns,optimization-techniques"

# Cross-agent learning
/agent-collaboration cross-learn
/agent-collaboration --agents "all" --knowledge-transfer --skill-sharing

# Best practices propagation
/agent-collaboration propagate-practices
/agent-collaboration --source security-agent --practices "secure-coding"
                     --targets "backend-agent,react-expert"

Collaboration Analytics

# Collaboration effectiveness
/agent-collaboration effectiveness
/agent-collaboration --collaboration-metrics --success-rates --efficiency

# Team dynamics analysis
/agent-collaboration dynamics
/agent-collaboration --interaction-patterns --communication-quality

# Collaborative output quality
/agent-collaboration quality
/agent-collaboration --joint-outputs --quality-metrics --improvements

Troubleshooting Agent Issues

Common Agent Problems

Agent Not Responding

Problem: Agent becomes unresponsive or fails to complete tasks

Diagnostic Steps:

# Check agent health
/agents health react-expert

# View agent logs
/debug agent react-expert --logs --recent

# Check resource usage
/agent-performance react-expert --resources

Solutions:

# Restart agent
/agents restart react-expert

# Reset agent state
/agents reset react-expert --preserve-learning

# Agent resource optimization
/optimize-agents react-expert --memory --context

Poor Agent Performance

Problem: Agent produces low-quality outputs or takes too long

Diagnostic Steps:

# Analyze performance metrics
/agent-performance react-expert --detailed

# Check recent tasks
/agents activity react-expert --recent --quality-analysis

# Review agent configuration
/agents info react-expert --config --capabilities

Solutions:

# Optimize agent configuration
/optimize-agents react-expert --auto-tune

# Update agent training
/agent-learning update react-expert --recent-feedback

# Adjust agent parameters
/agents configure react-expert --temperature 0.3 --max-tokens 3000

Agent Conflicts

Problem: Multiple agents produce conflicting outputs or approaches

Diagnostic Steps:

# Analyze conflict patterns
/orchestrate conflicts --agents "react-expert,universal-agent"

# Review collaboration history
/agent-collaboration history --conflicts --resolution

Solutions:

# Configure conflict resolution
/orchestrate configure --conflict-resolution expert-priority

# Improve agent coordination
/agent-collaboration optimize --coordination-patterns

# Establish agent hierarchies
/agents hierarchy --senior react-expert --junior universal-agent

Agent Debugging

Debug Mode

# Enable agent debugging
/debug agent react-expert --enable --verbose

# Debug specific agent action
/debug agent-action react-expert "create component" --trace

# Debug agent decision making
/debug agent-reasoning react-expert --decision-tree --context-analysis

Next Steps

Session Management

Learn about session control and history management

Configuration

Master NikCLI configuration and customization

Universal Agent

Deep dive into the Universal Agent capabilities

Agent Orchestration

Explore advanced agent orchestration patterns

Use /agents as your first command to understand available agents and their current status. The agent system is designed to be self-managing, but understanding agent capabilities helps you choose the right agent for each task.