[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72WP1C52HSCMP69VZC61ZB2":3,"skill-same:01M72WP1C52HSCMP69VZC61ZB2":46},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M72WP1C52HSCMP69VZC61ZB2","safety-interlocks",null,"Implement safety interlocks and protective mechanisms to prevent equipment damage and ensure safe control system operation.","agentscope-ai\u002FPawBench","agentscope-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F211762292?v=4","01M6CRT6543BZGGVK3P85GC08Q","AgentScope-AI","01M72WJAXCYDAXK74090327ER7",127,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fsafety-interlocks\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: safety-interlocks\ndescription: Implement safety interlocks and protective mechanisms to prevent equipment damage and ensure safe control system operation.\n---\n\n# Safety Interlocks for Control Systems\n\n## Overview\n\nSafety interlocks are protective mechanisms that prevent equipment damage and ensure safe operation. In control systems, the primary risks are output saturation and exceeding safe operating limits.\n\n## Implementation Pattern\n\nAlways check safety conditions BEFORE applying control outputs:\n\n```python\ndef apply_safety_limits(measurement, command, max_limit, min_limit, max_output, min_output):\n    \"\"\"\n    Apply safety checks and return safe command.\n\n    Args:\n        measurement: Current sensor reading\n        command: Requested control output\n        max_limit: Maximum safe measurement value\n        min_limit: Minimum safe measurement value\n        max_output: Maximum output command\n        min_output: Minimum output command\n\n    Returns:\n        tuple: (safe_command, safety_triggered)\n    \"\"\"\n    safety_triggered = False\n\n    # Check for over-limit - HIGHEST PRIORITY\n    if measurement >= max_limit:\n        command = min_output  # Emergency cutoff\n        safety_triggered = True\n\n    # Clamp output to valid range\n    command = max(min_output, min(max_output, command))\n\n    return command, safety_triggered\n```\n\n## Integration with Control Loop\n\n```python\nclass SafeController:\n    def __init__(self, controller, max_limit, min_output=0.0, max_output=100.0):\n        self.controller = controller\n        self.max_limit = max_limit\n        self.min_output = min_output\n        self.max_output = max_output\n        self.safety_events = []\n\n    def compute(self, measurement, dt):\n        \"\"\"Compute safe control output.\"\"\"\n        # Check safety FIRST\n        if measurement >= self.max_limit:\n            self.safety_events.append({\n                \"measurement\": measurement,\n                \"action\": \"emergency_cutoff\"\n            })\n            return self.min_output\n\n        # Normal control\n        output = self.controller.compute(measurement, dt)\n\n        # Clamp to valid range\n        return max(self.min_output, min(self.max_output, output))\n```\n\n## Safety During Open-Loop Testing\n\nDuring calibration\u002Fexcitation, safety is especially important because there's no feedback control:\n\n```python\ndef run_test_with_safety(system, input_value, duration, dt, max_limit):\n    \"\"\"Run open-loop test while monitoring safety limits.\"\"\"\n    data = []\n    current_input = input_value\n\n    for step in range(int(duration \u002F dt)):\n        result = system.step(current_input)\n        data.append(result)\n\n        # Safety check\n        if result[\"output\"] >= max_limit:\n            current_input = 0.0  # Cut input\n\n    return data\n```\n\n## Logging Safety Events\n\nAlways log safety events for analysis:\n\n```python\nsafety_log = {\n    \"limit\": max_limit,\n    \"events\": []\n}\n\nif measurement >= max_limit:\n    safety_log[\"events\"].append({\n        \"time\": current_time,\n        \"measurement\": measurement,\n        \"command_before\": command,\n        \"command_after\": 0.0,\n        \"event_type\": \"limit_exceeded\"\n    })\n```\n\n## Pre-Control Checklist\n\nBefore starting any control operation:\n\n1. **Verify sensor reading is reasonable**\n   - Not NaN or infinite\n   - Within physical bounds\n\n2. **Check initial conditions**\n   - Measurement should be at expected starting point\n   - Output should start at safe value\n\n3. **Confirm safety limits are configured**\n   - Maximum limit threshold set\n   - Output clamping enabled\n\n```python\ndef pre_control_checks(measurement, config):\n    \"\"\"Run pre-control safety verification.\"\"\"\n    assert not np.isnan(measurement), \"Measurement is NaN\"\n    assert config.get(\"max_limit\") is not None, \"Safety limit not configured\"\n    return True\n```\n\n## Best Practices\n\n1. **Defense in depth**: Multiple layers of protection\n2. **Fail safe**: When in doubt, reduce output\n3. **Log everything**: Record all safety events\n4. **Never bypass**: Safety code should not be conditionally disabled\n5. **Test safety**: Verify interlocks work before normal operation\n","# Safety Interlocks for Control Systems\n\n## Overview\n\nSafety interlocks are protective mechanisms that prevent equipment damage and ensure safe operation. In control systems, the primary risks are output saturation and exceeding safe operating limits.\n\n## Implementation Pattern\n\nAlways check safety conditions BEFORE applying control outputs:\n\n```python\ndef apply_safety_limits(measurement, command, max_limit, min_limit, max_output, min_output):\n    \"\"\"\n    Apply safety checks and return safe command.\n\n    Args:\n        measurement: Current sensor reading\n        command: Requested control output\n        max_limit: Maximum safe measurement value\n        min_limit: Minimum safe measurement value\n        max_output: Maximum output command\n        min_output: Minimum output command\n\n    Returns:\n        tuple: (safe_command, safety_triggered)\n    \"\"\"\n    safety_triggered = False\n\n    # Check for over-limit - HIGHEST PRIORITY\n    if measurement >= max_limit:\n        command = min_output  # Emergency cutoff\n        safety_triggered = True\n\n    # Clamp output to valid range\n    command = max(min_output, min(max_output, command))\n\n    return command, safety_triggered\n```\n\n## Integration with Control Loop\n\n```python\nclass SafeController:\n    def __init__(self, controller, max_limit, min_output=0.0, max_output=100.0):\n        self.controller = controller\n        self.max_limit = max_limit\n        self.min_output = min_output\n        self.max_output = max_output\n        self.safety_events = []\n\n    def compute(self, measurement, dt):\n        \"\"\"Compute safe control output.\"\"\"\n        # Check safety FIRST\n        if measurement >= self.max_limit:\n            self.safety_events.append({\n                \"measurement\": measurement,\n                \"action\": \"emergency_cutoff\"\n            })\n            return self.min_output\n\n        # Normal control\n        output = self.controller.compute(measurement, dt)\n\n        # Clamp to valid range\n        return max(self.min_output, min(self.max_output, output))\n```\n\n## Safety During Open-Loop Testing\n\nDuring calibration\u002Fexcitation, safety is especially important because there's no feedback control:\n\n```python\ndef run_test_with_safety(system, input_value, duration, dt, max_limit):\n    \"\"\"Run open-loop test while monitoring safety limits.\"\"\"\n    data = []\n    current_input = input_value\n\n    for step in range(int(duration \u002F dt)):\n        result = system.step(current_input)\n        data.append(result)\n\n        # Safety check\n        if result[\"output\"] >= max_limit:\n            current_input = 0.0  # Cut input\n\n    return data\n```\n\n## Logging Safety Events\n\nAlways log safety events for analysis:\n\n```python\nsafety_log = {\n    \"limit\": max_limit,\n    \"events\": []\n}\n\nif measurement >= max_limit:\n    safety_log[\"events\"].append({\n        \"time\": current_time,\n        \"measurement\": measurement,\n        \"command_before\": command,\n        \"command_after\": 0.0,\n        \"event_type\": \"limit_exceeded\"\n    })\n```\n\n## Pre-Control Checklist\n\nBefore starting any control operation:\n\n1. **Verify sensor reading is reasonable**\n   - Not NaN or infinite\n   - Within physical bounds\n\n2. **Check initial conditions**\n   - Measurement should be at expected starting point\n   - Output should start at safe value\n\n3. **Confirm safety limits are configured**\n   - Maximum limit threshold set\n   - Output clamping enabled\n\n```python\ndef pre_control_checks(measurement, config):\n    \"\"\"Run pre-control safety verification.\"\"\"\n    assert not np.isnan(measurement), \"Measurement is NaN\"\n    assert config.get(\"max_limit\") is not None, \"Safety limit not configured\"\n    return True\n```\n\n## Best Practices\n\n1. **Defense in depth**: Multiple layers of protection\n2. **Fail safe**: When in doubt, reduce output\n3. **Log everything**: Record all safety events\n4. **Never bypass**: Safety code should not be conditionally disabled\n5. **Test safety**: Verify interlocks work before normal operation\n",{"name":5,"description":7},"main","0f794a8bb6c27aa9ee4091b2691fa30e4ed9cc8f","data\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fsafety-interlocks",24,"https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench","https:\u002F\u002Fagentscope-ai.github.io\u002FPawBench\u002F","https:\u002F\u002Fgithub.com\u002Fagentscope-ai","2026-08-03 20:09:53",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add agentscope-ai\u002FPawBench --skill safety-interlocks",{"tool":38,"command":39},"bunx","bunx skills add agentscope-ai\u002FPawBench --skill safety-interlocks",{"tool":41,"command":42},"pnpm","pnpm dlx skills add agentscope-ai\u002FPawBench --skill safety-interlocks","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fsafety-interlocks\u002FSKILL.md",true,[],[47,53,60,67,74,81],{"skillNo":48,"name":49,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M72WQCJKBJZB6SC8MMYK3C8R","data-reconciliation","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT140_skillsbench_xlsx-recover-data\u002Fskills\u002Fdata-reconciliation\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M72WQ8R4DZKK1CT397C0ARQD","hierarchical-taxonomy-clustering","Build unified multi-level category taxonomy from hierarchical product category paths from any e-commerce companies using embedding-based recursive clustering with intelligent category naming via weighted word frequency analysis.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT139_skillsbench_taxonomy-tree-merge\u002Fskills\u002Fhierarchical-taxonomy-clustering\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M72WQ1989ZX613G3PY8Z492J","nlp-research-repo-package-installment","Align Python version and repo-declared dependencies (requirements.txt \u002F environment.yml) before installing packages for NLP research code reproduction.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT138_skillsbench_simpo-code-reproduction\u002Fskills\u002Fnlp-research-repo-package-installment\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M72WPZ828E746PAB3WCCQARW","state-space-linearization","Linearizing nonlinear dynamics around operating points for control design.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fstate-space-linearization\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M72WPXFXKWDFVTGAYX0B1TSA","mpc-horizon-tuning","Selecting MPC prediction horizon and cost matrices for web handling.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fmpc-horizon-tuning\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M72WPW3863555HM0CACRCCFK","integral-action-design","Adding integral action to MPC for offset-free tension tracking.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fintegral-action-design\u002FSKILL.md",[],[]]