refactor: clean up skills directory structure - move non-skill files to appropriate locations

Skills Directory Cleanup:
 NON-SKILL FILES MOVED: Proper directory organization
- .windsurf/meta/: Moved REFACTORING_SUMMARY.md and SKILL_ANALYSIS.md from skills/
- .windsurf/templates/: Moved agent-templates.md and workflow-templates.md from skills/openclaw-aitbc/
- .windsurf/references/: Moved ai-operations-reference.md from skills/openclaw-aitbc/
- scripts/: Moved setup.sh from skills/openclaw-aitbc/

 DEPRECATED SKILLS ARCHIVED: Clean skills directory structure
- .windsurf/skills/archive/: Moved aitbc-blockchain.md, openclaw-aitbc.md, openclaw-management.md
- These were legacy monolithic skills replaced by atomic skills
- Archive preserves history while keeping skills directory clean

 SKILLS DIRECTORY NOW CONTAINS: Only atomic, production-ready skills
- aitbc-ai-operator.md: AI job submission and monitoring
- aitbc-marketplace-participant.md: Marketplace operations and pricing
- aitbc-transaction-processor.md: Transaction execution and tracking
- aitbc-wallet-manager.md: Wallet creation, listing, balance checking
- openclaw-agent-communicator.md: Agent message handling and responses
- openclaw-session-manager.md: Session creation and context management
- archive/: Deprecated legacy skills (3 files)

DIRECTORY STRUCTURE IMPROVEMENT:
🎯 Skills Directory: Contains only 6 atomic skills + archive
📋 Meta Directory: Contains refactoring analysis and summaries
📝 Templates Directory: Contains agent and workflow templates
📖 References Directory: Contains reference documentation and guides
🗂️ Archive Directory: Contains deprecated legacy skills

BENEFITS:
 Clean Skills Directory: Only contains actual atomic skills
 Proper Organization: Non-skill files in appropriate directories
 Archive Preservation: Legacy skills preserved for reference
 Maintainability: Clear separation of concerns
 Navigation: Easier to find and use actual skills

Result: Skills directory now properly organized with only atomic skills, non-skill files moved to appropriate locations, and deprecated skills archived for reference.
This commit is contained in:
2026-03-30 17:05:12 +02:00
parent 6d5bc30d87
commit fa2b90b094
10 changed files with 130 additions and 342 deletions

View File

@@ -0,0 +1,490 @@
---
description: Complete AITBC blockchain operations and integration
title: AITBC Blockchain Operations Skill
version: 1.0
---
# AITBC Blockchain Operations Skill
This skill provides comprehensive AITBC blockchain operations including wallet management, transactions, AI operations, marketplace participation, and node coordination.
## Prerequisites
- AITBC multi-node blockchain operational (aitbc genesis, aitbc1 follower)
- AITBC CLI accessible: `/opt/aitbc/aitbc-cli`
- SSH access between nodes for cross-node operations
- Systemd services: `aitbc-blockchain-node.service`, `aitbc-blockchain-rpc.service`
- Poetry 2.3.3+ for Python package management
- Wallet passwords known (default: 123 for new wallets)
## Critical: Correct CLI Syntax
### AITBC CLI Commands
```bash
# All commands run from /opt/aitbc with venv active
cd /opt/aitbc && source venv/bin/activate
# Basic Operations
./aitbc-cli create --name wallet-name # Create wallet
./aitbc-cli list # List wallets
./aitbc-cli balance --name wallet-name # Check balance
./aitbc-cli send --from w1 --to addr --amount 100 --password pass
./aitbc-cli chain # Blockchain info
./aitbc-cli network # Network status
./aitbc-cli analytics # Analytics data
```
### Cross-Node Operations
```bash
# Always activate venv on remote nodes
ssh aitbc1 'cd /opt/aitbc && source venv/bin/activate && ./aitbc-cli list'
# Cross-node transaction
./aitbc-cli send --from genesis-ops --to ait141b3bae6eea3a74273ef3961861ee58e12b6d855 --amount 100 --password 123
```
## Wallet Management
### Creating Wallets
```bash
# Create new wallet with password
./aitbc-cli create --name my-wallet --password 123
# List all wallets
./aitbc-cli list
# Check wallet balance
./aitbc-cli balance --name my-wallet
```
### Wallet Operations
```bash
# Send transaction
./aitbc-cli send --from wallet1 --to wallet2 --amount 100 --password 123
# Check transaction history
./aitbc-cli transactions --name my-wallet
# Import wallet from keystore
./aitbc-cli import --keystore /path/to/keystore.json --password 123
```
### Standard Wallet Addresses
```bash
# Genesis operations wallet
./aitbc-cli balance --name genesis-ops
# Address: ait158ec7a0713f30ccfb1aac6bfbab71f36271c5871
# Follower operations wallet
./aitbc-cli balance --name follower-ops
# Address: ait141b3bae6eea3a74273ef3961861ee58e12b6d855
```
## Blockchain Operations
### Chain Information
```bash
# Get blockchain status
./aitbc-cli chain
# Get network status
./aitbc-cli network
# Get analytics data
./aitbc-cli analytics
# Check block height
curl -s http://localhost:8006/rpc/head | jq .height
```
### Node Status
```bash
# Check health endpoint
curl -s http://localhost:8006/health | jq .
# Check both nodes
curl -s http://localhost:8006/health | jq .
ssh aitbc1 'curl -s http://localhost:8006/health | jq .'
# Check services
systemctl is-active aitbc-blockchain-node.service aitbc-blockchain-rpc.service
ssh aitbc1 'systemctl is-active aitbc-blockchain-node.service aitbc-blockchain-rpc.service'
```
### Synchronization Monitoring
```bash
# Check height difference
GENESIS_HEIGHT=$(curl -s http://localhost:8006/rpc/head | jq .height)
FOLLOWER_HEIGHT=$(ssh aitbc1 'curl -s http://localhost:8006/rpc/head | jq .height')
echo "Height diff: $((FOLLOWER_HEIGHT - GENESIS_HEIGHT))"
# Comprehensive health check
python3 /tmp/aitbc1_heartbeat.py
```
## Agent Operations
### Creating Agents
```bash
# Create basic agent
./aitbc-cli agent create --name agent-name --description "Agent description"
# Create agent with full verification
./aitbc-cli agent create --name agent-name --description "Agent description" --verification full
# Create AI-specific agent
./aitbc-cli agent create --name ai-agent --description "AI processing agent" --verification full
```
### Managing Agents
```bash
# Execute agent
./aitbc-cli agent execute --name agent-name --wallet wallet --priority high
# Check agent status
./aitbc-cli agent status --name agent-name
# List all agents
./aitbc-cli agent list
```
## AI Operations
### AI Job Submission
```bash
# Inference job
./aitbc-cli ai-submit --wallet genesis-ops --type inference --prompt "Generate image" --payment 100
# Training job
./aitbc-cli ai-submit --wallet genesis-ops --type training --model "gpt-3.5" --dataset "data.json" --payment 500
# Multimodal job
./aitbc-cli ai-submit --wallet genesis-ops --type multimodal --prompt "Analyze image" --image-path "/path/to/img.jpg" --payment 200
```
### AI Job Types
- **inference**: Image generation, text analysis, predictions
- **training**: Model training on datasets
- **processing**: Data transformation and analysis
- **multimodal**: Combined text, image, audio processing
### AI Job Monitoring
```bash
# Check job status
./aitbc-cli ai-status --job-id job_123
# Check job history
./aitbc-cli ai-history --wallet genesis-ops --limit 10
# Estimate job cost
./aitbc-cli ai-estimate --type inference --prompt-length 100 --resolution 512
```
## Resource Management
### Resource Allocation
```bash
# Allocate GPU resources
./aitbc-cli resource allocate --agent-id ai-agent --gpu 1 --memory 8192 --duration 3600
# Allocate CPU resources
./aitbc-cli resource allocate --agent-id data-processor --cpu 4 --memory 4096 --duration 1800
# Check resource status
./aitbc-cli resource status
# List allocated resources
./aitbc-cli resource list
```
### Resource Types
- **gpu**: GPU units for AI inference
- **cpu**: CPU cores for processing
- **memory**: RAM in megabytes
- **duration**: Reservation time in seconds
## Marketplace Operations
### Creating Services
```bash
# Create AI service
./aitbc-cli marketplace --action create --name "AI Image Generation" --type ai-inference --price 50 --wallet genesis-ops --description "Generate high-quality images"
# Create training service
./aitbc-cli marketplace --action create --name "Model Training" --type ai-training --price 200 --wallet genesis-ops --description "Train custom models"
# Create data processing service
./aitbc-cli marketplace --action create --name "Data Analysis" --type ai-processing --price 75 --wallet genesis-ops --description "Analyze datasets"
```
### Marketplace Interaction
```bash
# List available services
./aitbc-cli marketplace --action list
# Search for services
./aitbc-cli marketplace --action search --query "AI"
# Bid on service
./aitbc-cli marketplace --action bid --service-id service_123 --amount 60 --wallet genesis-ops
# Execute purchased service
./aitbc-cli marketplace --action execute --service-id service_123 --job-data "prompt:Generate landscape image"
# Check my listings
./aitbc-cli marketplace --action my-listings --wallet genesis-ops
```
## Mining Operations
### Mining Control
```bash
# Start mining
./aitbc-cli mine-start --wallet genesis-ops
# Stop mining
./aitbc-cli mine-stop
# Check mining status
./aitbc-cli mine-status
```
## Smart Contract Messaging
### Topic Management
```bash
# Create coordination topic
curl -X POST http://localhost:8006/rpc/messaging/topics/create \
-H "Content-Type: application/json" \
-d '{"agent_id": "agent", "agent_address": "address", "title": "Topic", "description": "Description", "tags": ["coordination"]}'
# List topics
curl -s http://localhost:8006/rpc/messaging/topics
# Get topic messages
curl -s http://localhost:8006/rpc/messaging/topics/topic_id/messages
```
### Message Operations
```bash
# Post message to topic
curl -X POST http://localhost:8006/rpc/messaging/messages/post \
-H "Content-Type: application/json" \
-d '{"agent_id": "agent", "agent_address": "address", "topic_id": "topic_id", "content": "Message content"}'
# Vote on message
curl -X POST http://localhost:8006/rpc/messaging/messages/message_id/vote \
-H "Content-Type: application/json" \
-d '{"agent_id": "agent", "agent_address": "address", "vote_type": "upvote"}'
# Check agent reputation
curl -s http://localhost:8006/rpc/messaging/agents/agent_id/reputation
```
## Cross-Node Coordination
### Cross-Node Transactions
```bash
# Send from genesis to follower
./aitbc-cli send --from genesis-ops --to ait141b3bae6eea3a74273ef3961861ee58e12b6d855 --amount 100 --password 123
# Send from follower to genesis
ssh aitbc1 'cd /opt/aitbc && source venv/bin/activate && ./aitbc-cli send --from follower-ops --to ait158ec7a0713f30ccfb1aac6bfbab71f36271c5871 --amount 50 --password 123'
```
### Cross-Node AI Operations
```bash
# Submit AI job to specific node
./aitbc-cli ai-submit --wallet genesis-ops --type inference --prompt "Generate image" --target-node "aitbc1" --payment 100
# Distribute training across nodes
./aitbc-cli ai-submit --wallet genesis-ops --type training --model "distributed-model" --nodes "aitbc,aitbc1" --payment 500
```
## Configuration Management
### Environment Configuration
```bash
# Check current configuration
cat /etc/aitbc/.env
# Key configuration parameters
chain_id=ait-mainnet
proposer_id=ait158ec7a0713f30ccfb1aac6bfbab71f36271c5871
enable_block_production=true
mempool_backend=database
gossip_backend=redis
gossip_broadcast_url=redis://10.1.223.40:6379
```
### Service Management
```bash
# Restart services
sudo systemctl restart aitbc-blockchain-node.service aitbc-blockchain-rpc.service
# Check service logs
sudo journalctl -u aitbc-blockchain-node.service -f
sudo journalctl -u aitbc-blockchain-rpc.service -f
# Cross-node service restart
ssh aitbc1 'sudo systemctl restart aitbc-blockchain-node.service aitbc-blockchain-rpc.service'
```
## Data Management
### Database Operations
```bash
# Check database files
ls -la /var/lib/aitbc/data/ait-mainnet/
# Backup database
sudo cp /var/lib/aitbc/data/ait-mainnet/chain.db /var/lib/aitbc/data/ait-mainnet/chain.db.backup.$(date +%s)
# Reset blockchain (genesis creation)
sudo systemctl stop aitbc-blockchain-node.service aitbc-blockchain-rpc.service
sudo mv /var/lib/aitbc/data/ait-mainnet/chain.db /var/lib/aitbc/data/ait-mainnet/chain.db.backup.$(date +%s)
sudo systemctl start aitbc-blockchain-node.service aitbc-blockchain-rpc.service
```
### Genesis Configuration
```bash
# Create genesis.json with allocations
cat << 'EOF' | sudo tee /var/lib/aitbc/data/ait-mainnet/genesis.json
{
"allocations": [
{
"address": "ait158ec7a0713f30ccfb1aac6bfbab71f36271c5871",
"balance": 1000000,
"nonce": 0
}
],
"authorities": [
{
"address": "ait158ec7a0713f30ccfb1aac6bfbab71f36271c5871",
"weight": 1
}
]
}
EOF
```
## Monitoring and Analytics
### Health Monitoring
```bash
# Comprehensive health check
python3 /tmp/aitbc1_heartbeat.py
# Manual health checks
curl -s http://localhost:8006/health | jq .
ssh aitbc1 'curl -s http://localhost:8006/health | jq .'
# Check sync status
./aitbc-cli chain
./aitbc-cli network
```
### Performance Metrics
```bash
# Check block production rate
watch -n 10 './aitbc-cli chain | grep "Height:"'
# Monitor transaction throughput
./aitbc-cli analytics
# Check resource utilization
./aitbc-cli resource status
```
## Troubleshooting
### Common Issues and Solutions
#### Transactions Not Mining
```bash
# Check proposer status
curl -s http://localhost:8006/health | jq .proposer_id
# Check mempool status
curl -s http://localhost:8006/rpc/mempool
# Verify mempool configuration
grep mempool_backend /etc/aitbc/.env
```
#### RPC Connection Issues
```bash
# Check RPC service
systemctl status aitbc-blockchain-rpc.service
# Test RPC endpoint
curl -s http://localhost:8006/health
# Check port availability
netstat -tlnp | grep 8006
```
#### Wallet Issues
```bash
# Check wallet exists
./aitbc-cli list | grep wallet-name
# Test wallet password
./aitbc-cli balance --name wallet-name --password 123
# Create new wallet if needed
./aitbc-cli create --name new-wallet --password 123
```
#### Sync Issues
```bash
# Check both nodes' heights
curl -s http://localhost:8006/rpc/head | jq .height
ssh aitbc1 'curl -s http://localhost:8006/rpc/head | jq .height'
# Check gossip connectivity
grep gossip_broadcast_url /etc/aitbc/.env
# Restart services if needed
sudo systemctl restart aitbc-blockchain-node.service
```
## Standardized Paths
| Resource | Path |
|---|---|
| Blockchain data | `/var/lib/aitbc/data/ait-mainnet/` |
| Keystore | `/var/lib/aitbc/keystore/` |
| Environment config | `/etc/aitbc/.env` |
| CLI tool | `/opt/aitbc/aitbc-cli` |
| Scripts | `/opt/aitbc/scripts/` |
| Logs | `/var/log/aitbc/` |
| Services | `/etc/systemd/system/aitbc-*.service` |
## Best Practices
### Security
- Use strong wallet passwords
- Keep keystore files secure
- Monitor transaction activity
- Use proper authentication for RPC endpoints
### Performance
- Monitor resource utilization
- Optimize transaction batching
- Use appropriate thinking levels for AI operations
- Regular database maintenance
### Operations
- Regular health checks
- Backup critical data
- Monitor cross-node synchronization
- Keep documentation updated
### Development
- Test on development network first
- Use proper version control
- Document all changes
- Implement proper error handling
This AITBC Blockchain Operations skill provides comprehensive coverage of all blockchain operations, from basic wallet management to advanced AI operations and cross-node coordination.

View File

@@ -0,0 +1,170 @@
---
description: Legacy OpenClaw AITBC integration - see split skills for focused operations
title: OpenClaw AITBC Integration (Legacy)
version: 6.0 - DEPRECATED
---
# OpenClaw AITBC Integration (Legacy - See Split Skills)
⚠️ **This skill has been split into focused skills for better organization:**
## 📚 New Split Skills
### 1. OpenClaw Agent Management Skill
**File**: `openclaw-management.md`
**Focus**: Pure OpenClaw agent operations, communication, and coordination
- Agent creation and management
- Session-based workflows
- Cross-agent communication
- Performance optimization
- Error handling and debugging
**Use for**: Agent orchestration, workflow coordination, multi-agent systems
### 2. AITBC Blockchain Operations Skill
**File**: `aitbc-blockchain.md`
**Focus**: Pure AITBC blockchain operations and integration
- Wallet management and transactions
- AI operations and marketplace
- Node coordination and monitoring
- Smart contract messaging
- Cross-node operations
**Use for**: Blockchain operations, AI jobs, marketplace participation, node management
## Migration Guide
### From Legacy to Split Skills
**Before (Legacy)**:
```bash
# Mixed OpenClaw + AITBC operations
openclaw agent --agent main --message "Check blockchain and process data" --thinking high
cd /opt/aitbc && source venv/bin/activate && ./aitbc-cli chain
```
**After (Split Skills)**:
**OpenClaw Agent Management**:
```bash
# Pure agent coordination
openclaw agent --agent coordinator --message "Coordinate blockchain monitoring workflow" --thinking high
# Agent workflow orchestration
SESSION_ID="blockchain-monitor-$(date +%s)"
openclaw agent --agent monitor --session-id $SESSION_ID --message "Monitor blockchain health" --thinking medium
```
**AITBC Blockchain Operations**:
```bash
# Pure blockchain operations
cd /opt/aitbc && source venv/bin/activate
./aitbc-cli chain
./aitbc-cli ai-submit --wallet genesis-ops --type inference --prompt "Generate image" --payment 100
```
## Why the Split?
### Benefits of Focused Skills
1. **Clearer Separation of Concerns**
- OpenClaw: Agent coordination and workflow management
- AITBC: Blockchain operations and data management
2. **Better Documentation Organization**
- Each skill focuses on its domain expertise
- Reduced cognitive load when learning
- Easier maintenance and updates
3. **Improved Reusability**
- OpenClaw skills can be used with any system
- AITBC skills can be used with any agent framework
- Modular combination possible
4. **Enhanced Searchability**
- Find relevant commands faster
- Domain-specific troubleshooting
- Focused best practices
### When to Use Each Skill
**Use OpenClaw Agent Management Skill for**:
- Multi-agent workflow coordination
- Agent communication patterns
- Session management and context
- Agent performance optimization
- Error handling and debugging
**Use AITBC Blockchain Operations Skill for**:
- Wallet and transaction management
- AI job submission and monitoring
- Marketplace operations
- Node health and synchronization
- Smart contract messaging
**Combine Both Skills for**:
- Complete OpenClaw + AITBC integration
- Agent-driven blockchain operations
- Automated blockchain workflows
- Cross-node agent coordination
## Legacy Content (Deprecated)
The following content from the original combined skill is now deprecated and moved to the appropriate split skills:
- ~~Agent command syntax~~ → **OpenClaw Agent Management**
- ~~AITBC CLI commands~~ → **AITBC Blockchain Operations**
- ~~AI operations~~ → **AITBC Blockchain Operations**
- ~~Blockchain coordination~~ → **AITBC Blockchain Operations**
- ~~Agent workflows~~ → **OpenClaw Agent Management**
## Migration Checklist
### ✅ Completed
- [x] Created OpenClaw Agent Management skill
- [x] Created AITBC Blockchain Operations skill
- [x] Updated all command references
- [x] Added migration guide
### 🔄 In Progress
- [ ] Update workflow scripts to use split skills
- [ ] Update documentation references
- [ ] Test split skills independently
### 📋 Next Steps
- [ ] Remove legacy content after validation
- [ ] Update integration examples
- [ ] Create combined usage examples
## Quick Reference
### OpenClaw Agent Management
```bash
# Agent coordination
openclaw agent --agent coordinator --message "Coordinate workflow" --thinking high
# Session-based workflow
SESSION_ID="task-$(date +%s)"
openclaw agent --agent worker --session-id $SESSION_ID --message "Execute task" --thinking medium
```
### AITBC Blockchain Operations
```bash
# Blockchain status
cd /opt/aitbc && source venv/bin/activate
./aitbc-cli chain
# AI operations
./aitbc-cli ai-submit --wallet wallet --type inference --prompt "Generate image" --payment 100
```
---
**Recommendation**: Use the new split skills for all new development. This legacy skill is maintained for backward compatibility but will be deprecated in future versions.
## Quick Links to New Skills
- **OpenClaw Agent Management**: [openclaw-management.md](openclaw-management.md)
- **AITBC Blockchain Operations**: [aitbc-blockchain.md](aitbc-blockchain.md)

View File

@@ -0,0 +1,344 @@
---
description: OpenClaw agent management and coordination capabilities
title: OpenClaw Agent Management Skill
version: 1.0
---
# OpenClaw Agent Management Skill
This skill provides comprehensive OpenClaw agent management, communication, and coordination capabilities. Focus on agent operations, session management, and cross-agent workflows.
## Prerequisites
- OpenClaw 2026.3.24+ installed and gateway running
- Agent workspace configured: `~/.openclaw/workspace/`
- Network connectivity for multi-agent coordination
## Critical: Correct OpenClaw Syntax
### Agent Commands
```bash
# CORRECT — always use --message (long form), not -m
openclaw agent --agent main --message "Your task here" --thinking medium
# Session-based communication (maintains context across calls)
SESSION_ID="workflow-$(date +%s)"
openclaw agent --agent main --session-id $SESSION_ID --message "Initialize task" --thinking low
openclaw agent --agent main --session-id $SESSION_ID --message "Continue task" --thinking medium
# Thinking levels: off | minimal | low | medium | high | xhigh
```
> **WARNING**: The `-m` short form does NOT work reliably. Always use `--message`.
> **WARNING**: `--session-id` is required to maintain conversation context across multiple agent calls.
### Agent Status and Management
```bash
# Check agent status
openclaw status --agent all
openclaw status --agent main
# List available agents
openclaw list --agents
# Agent workspace management
openclaw workspace --setup
openclaw workspace --status
```
## Agent Communication Patterns
### Single Agent Tasks
```bash
# Simple task execution
openclaw agent --agent main --message "Analyze the system logs and report any errors" --thinking high
# Task with specific parameters
openclaw agent --agent main --message "Process this data: /path/to/data.csv" --thinking medium --parameters "format:csv,mode:analyze"
```
### Session-Based Workflows
```bash
# Initialize session
SESSION_ID="data-analysis-$(date +%s)"
# Step 1: Data collection
openclaw agent --agent main --session-id $SESSION_ID --message "Collect data from API endpoints" --thinking low
# Step 2: Data processing
openclaw agent --agent main --session-id $SESSION_ID --message "Process collected data and generate insights" --thinking medium
# Step 3: Report generation
openclaw agent --agent main --session-id $SESSION_ID --message "Create comprehensive report with visualizations" --thinking high
```
### Multi-Agent Coordination
```bash
# Coordinator agent manages workflow
openclaw agent --agent coordinator --message "Coordinate data processing across multiple agents" --thinking high
# Worker agents execute specific tasks
openclaw agent --agent worker-1 --message "Process dataset A" --thinking medium
openclaw agent --agent worker-2 --message "Process dataset B" --thinking medium
# Aggregator combines results
openclaw agent --agent aggregator --message "Combine results from worker-1 and worker-2" --thinking high
```
## Agent Types and Roles
### Coordinator Agent
```bash
# Setup coordinator for complex workflows
openclaw agent --agent coordinator --message "Initialize as workflow coordinator. Manage task distribution, monitor progress, aggregate results." --thinking high
# Use coordinator for orchestration
openclaw agent --agent coordinator --message "Orchestrate data pipeline: extract → transform → load → validate" --thinking high
```
### Worker Agent
```bash
# Setup worker for specific tasks
openclaw agent --agent worker --message "Initialize as data processing worker. Execute assigned tasks efficiently." --thinking medium
# Assign specific work
openclaw agent --agent worker --message "Process customer data file: /data/customers.json" --thinking medium
```
### Monitor Agent
```bash
# Setup monitor for oversight
openclaw agent --agent monitor --message "Initialize as system monitor. Track performance, detect anomalies, report status." --thinking low
# Continuous monitoring
openclaw agent --agent monitor --message "Monitor system health and report any issues" --thinking minimal
```
## Agent Workflows
### Data Processing Workflow
```bash
SESSION_ID="data-pipeline-$(date +%s)"
# Phase 1: Data Extraction
openclaw agent --agent extractor --session-id $SESSION_ID --message "Extract data from sources" --thinking medium
# Phase 2: Data Transformation
openclaw agent --agent transformer --session-id $SESSION_ID --message "Transform extracted data" --thinking medium
# Phase 3: Data Loading
openclaw agent --agent loader --session-id $SESSION_ID --message "Load transformed data to destination" --thinking medium
# Phase 4: Validation
openclaw agent --agent validator --session-id $SESSION_ID --message "Validate loaded data integrity" --thinking high
```
### Monitoring Workflow
```bash
SESSION_ID="monitoring-$(date +%s)"
# Continuous monitoring loop
while true; do
openclaw agent --agent monitor --session-id $SESSION_ID --message "Check system health" --thinking minimal
sleep 300 # Check every 5 minutes
done
```
### Analysis Workflow
```bash
SESSION_ID="analysis-$(date +%s)"
# Initial analysis
openclaw agent --agent analyst --session-id $SESSION_ID --message "Perform initial data analysis" --thinking high
# Deep dive analysis
openclaw agent --agent analyst --session-id $SESSION_ID --message "Deep dive into anomalies and patterns" --thinking high
# Report generation
openclaw agent --agent analyst --session-id $SESSION_ID --message "Generate comprehensive analysis report" --thinking high
```
## Agent Configuration
### Agent Parameters
```bash
# Agent with specific parameters
openclaw agent --agent main --message "Process data" --thinking medium \
--parameters "input_format:json,output_format:csv,mode:batch"
# Agent with timeout
openclaw agent --agent main --message "Long running task" --thinking high \
--parameters "timeout:3600,retry_count:3"
# Agent with resource constraints
openclaw agent --agent main --message "Resource-intensive task" --thinking high \
--parameters "max_memory:4GB,max_cpu:2,max_duration:1800"
```
### Agent Context Management
```bash
# Set initial context
openclaw agent --agent main --message "Initialize with context: data_analysis_v2" --thinking low \
--context "project:data_analysis,version:2.0,dataset:customer_data"
# Maintain context across calls
openclaw agent --agent main --session-id $SESSION_ID --message "Continue with previous context" --thinking medium
# Update context
openclaw agent --agent main --session-id $SESSION_ID --message "Update context: new_phase" --thinking medium \
--context-update "phase:processing,status:active"
```
## Agent Communication
### Cross-Agent Messaging
```bash
# Agent A sends message to Agent B
openclaw agent --agent agent-a --message "Send results to agent-b" --thinking medium \
--send-to "agent-b" --message-type "results"
# Agent B receives and processes
openclaw agent --agent agent-b --message "Process received results" --thinking medium \
--receive-from "agent-a"
```
### Agent Collaboration
```bash
# Setup collaboration team
TEAM_ID="team-analytics-$(date +%s)"
# Team leader coordination
openclaw agent --agent team-lead --session-id $TEAM_ID --message "Coordinate team analytics workflow" --thinking high
# Team member tasks
openclaw agent --agent analyst-1 --session-id $TEAM_ID --message "Analyze customer segment A" --thinking high
openclaw agent --agent analyst-2 --session-id $TEAM_ID --message "Analyze customer segment B" --thinking high
# Team consolidation
openclaw agent --agent team-lead --session-id $TEAM_ID --message "Consolidate team analysis results" --thinking high
```
## Agent Error Handling
### Error Recovery
```bash
# Agent with error handling
openclaw agent --agent main --message "Process data with error handling" --thinking medium \
--parameters "error_handling:retry_on_failure,max_retries:3,fallback_mode:graceful_degradation"
# Monitor agent errors
openclaw agent --agent monitor --message "Check for agent errors and report" --thinking low \
--parameters "check_type:error_log,alert_threshold:5"
```
### Agent Debugging
```bash
# Debug mode
openclaw agent --agent main --message "Debug task execution" --thinking high \
--parameters "debug:true,log_level:verbose,trace_execution:true"
# Agent state inspection
openclaw agent --agent main --message "Report current state and context" --thinking low \
--parameters "report_type:state,include_context:true"
```
## Agent Performance Optimization
### Efficient Agent Usage
```bash
# Batch processing
openclaw agent --agent processor --message "Process data in batches" --thinking medium \
--parameters "batch_size:100,parallel_processing:true"
# Resource optimization
openclaw agent --agent optimizer --message "Optimize resource usage" --thinking high \
--parameters "memory_efficiency:true,cpu_optimization:true"
```
### Agent Scaling
```bash
# Scale out work
for i in {1..5}; do
openclaw agent --agent worker-$i --message "Process batch $i" --thinking medium &
done
# Scale in coordination
openclaw agent --agent coordinator --message "Coordinate scaled-out workers" --thinking high
```
## Agent Security
### Secure Agent Operations
```bash
# Agent with security constraints
openclaw agent --agent secure-agent --message "Process sensitive data" --thinking high \
--parameters "security_level:high,data_encryption:true,access_log:true"
# Agent authentication
openclaw agent --agent authenticated-agent --message "Authenticated operation" --thinking medium \
--parameters "auth_required:true,token_expiry:3600"
```
## Agent Monitoring and Analytics
### Performance Monitoring
```bash
# Monitor agent performance
openclaw agent --agent monitor --message "Monitor agent performance metrics" --thinking low \
--parameters "metrics:cpu,memory,tasks_per_second,error_rate"
# Agent analytics
openclaw agent --agent analytics --message "Generate agent performance report" --thinking medium \
--parameters "report_type:performance,period:last_24h"
```
## Troubleshooting Agent Issues
### Common Agent Problems
1. **Session Loss**: Use consistent `--session-id` across calls
2. **Context Loss**: Maintain context with `--context` parameter
3. **Performance Issues**: Optimize `--thinking` level and task complexity
4. **Communication Failures**: Check agent status and network connectivity
### Debug Commands
```bash
# Check agent status
openclaw status --agent all
# Test agent communication
openclaw agent --agent main --message "Ping test" --thinking minimal
# Check workspace
openclaw workspace --status
# Verify agent configuration
openclaw config --show --agent main
```
## Best Practices
### Session Management
- Use meaningful session IDs: `task-type-$(date +%s)`
- Maintain context across related tasks
- Clean up sessions when workflows complete
### Thinking Level Optimization
- **off**: Simple, repetitive tasks
- **minimal**: Quick status checks, basic operations
- **low**: Data processing, routine analysis
- **medium**: Complex analysis, decision making
- **high**: Strategic planning, complex problem solving
- **xhigh**: Critical decisions, creative tasks
### Agent Organization
- Use descriptive agent names: `data-processor`, `monitor`, `coordinator`
- Group related agents in workflows
- Implement proper error handling and recovery
### Performance Tips
- Batch similar operations
- Use appropriate thinking levels
- Monitor agent resource usage
- Implement proper session cleanup
This OpenClaw Agent Management skill provides the foundation for effective agent coordination, communication, and workflow orchestration across any domain or application.