aitbc/.windsurf/skills/openclaw-agent-testing-skill.md

description, title, version

description	title	version
Atomic OpenClaw agent testing with deterministic communication validation and performance metrics	openclaw-agent-testing-skill	1.0

OpenClaw Agent Testing Skill

Purpose

Test and validate OpenClaw agent functionality, communication patterns, session management, and performance with deterministic validation metrics.

Activation

Trigger when user requests OpenClaw agent testing: agent functionality validation, communication testing, session management testing, or agent performance analysis.

Input

{
  "operation": "test-agent-communication|test-session-management|test-agent-performance|test-multi-agent|comprehensive",
  "agent": "main|specific_agent_name (default: main)",
  "test_message": "string (optional for communication testing)",
  "session_id": "string (optional for session testing)",
  "thinking_level": "off|minimal|low|medium|high|xhigh",
  "test_duration": "number (optional, default: 60 seconds)",
  "message_count": "number (optional, default: 5)",
  "concurrent_agents": "number (optional, default: 2)"
}

Output

{
  "summary": "OpenClaw agent testing completed successfully",
  "operation": "test-agent-communication|test-session-management|test-agent-performance|test-multi-agent|comprehensive",
  "test_results": {
    "agent_communication": "boolean",
    "session_management": "boolean",
    "agent_performance": "boolean",
    "multi_agent_coordination": "boolean"
  },
  "agent_details": {
    "agent_name": "string",
    "agent_status": "online|offline|error",
    "response_time": "number",
    "message_success_rate": "number"
  },
  "communication_metrics": {
    "messages_sent": "number",
    "messages_received": "number",
    "average_response_time": "number",
    "communication_success_rate": "number"
  },
  "session_metrics": {
    "sessions_created": "number",
    "session_preservation": "boolean",
    "context_maintenance": "boolean",
    "session_duration": "number"
  },
  "performance_metrics": {
    "cpu_usage": "number",
    "memory_usage": "number",
    "response_latency": "number",
    "throughput": "number"
  },
  "issues": [],
  "recommendations": [],
  "confidence": 1.0,
  "execution_time": "number",
  "validation_status": "success|partial|failed"
}

Process

1. Analyze

• Validate agent testing parameters and operation type
• Check OpenClaw service availability and health
• Verify agent availability and status
• Assess testing scope and requirements

2. Plan

• Prepare agent communication test scenarios
• Define session management testing strategy
• Set performance monitoring and validation criteria
• Configure multi-agent coordination tests

3. Execute

• Test agent communication with various thinking levels
• Validate session creation and context preservation
• Monitor agent performance and resource utilization
• Test multi-agent coordination and communication patterns

4. Validate

• Verify agent communication success and response quality
• Check session management effectiveness and context preservation
• Validate agent performance metrics and resource usage
• Confirm multi-agent coordination and communication patterns

Constraints

• MUST NOT test unavailable agents without explicit request
• MUST NOT exceed message length limits (4000 characters)
• MUST validate thinking level compatibility
• MUST handle communication timeouts gracefully
• MUST preserve session context during testing
• MUST provide deterministic performance metrics

Environment Assumptions

• OpenClaw 2026.3.24+ installed and gateway running
• Agent workspace configured at ~/.openclaw/workspace/
• Network connectivity for agent communication
• Default agent available: "main"
• Session management functional

Error Handling

• Agent unavailable → Return agent status and availability recommendations
• Communication timeout → Return timeout details and retry suggestions
• Session management failures → Return session diagnostics and recovery steps
• Performance issues → Return performance metrics and optimization recommendations

Example Usage Prompt

Run comprehensive OpenClaw agent testing including communication, session management, performance, and multi-agent coordination validation

Expected Output Example

{
  "summary": "Comprehensive OpenClaw agent testing completed with all systems operational",
  "operation": "comprehensive",
  "test_results": {
    "agent_communication": true,
    "session_management": true,
    "agent_performance": true,
    "multi_agent_coordination": true
  },
  "agent_details": {
    "agent_name": "main",
    "agent_status": "online",
    "response_time": 2.3,
    "message_success_rate": 100.0
  },
  "communication_metrics": {
    "messages_sent": 5,
    "messages_received": 5,
    "average_response_time": 2.1,
    "communication_success_rate": 100.0
  },
  "session_metrics": {
    "sessions_created": 3,
    "session_preservation": true,
    "context_maintenance": true,
    "session_duration": 45.2
  },
  "performance_metrics": {
    "cpu_usage": 15.3,
    "memory_usage": 85.2,
    "response_latency": 2.1,
    "throughput": 2.4
  },
  "issues": [],
  "recommendations": ["All agents operational", "Communication latency optimal", "Session management effective"],
  "confidence": 1.0,
  "execution_time": 67.3,
  "validation_status": "success"
}

Model Routing Suggestion

Fast Model (Claude Haiku, GPT-3.5-turbo)

• Simple agent availability checking
• Basic communication testing with low thinking
• Quick agent status validation

Reasoning Model (Claude Sonnet, GPT-4)

• Comprehensive agent communication testing
• Session management validation and optimization
• Multi-agent coordination testing and analysis
• Complex agent performance diagnostics

Coding Model (Claude Sonnet, GPT-4)

• Agent performance optimization algorithms
• Communication pattern analysis and improvement
• Session management enhancement strategies

Performance Notes

• Execution Time: 5-15 seconds for basic tests, 30-90 seconds for comprehensive testing
• Memory Usage: <150MB for agent testing operations
• Network Requirements: OpenClaw gateway connectivity
• Concurrency: Safe for multiple simultaneous agent tests with different agents
• Session Management: Automatic session creation and context preservation testing