[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6EHJ27QWA786484SWXW58HN":3,"skill-same:01M6EHJ27QWA786484SWXW58HN":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":43,"commitSha":44,"skillDirectory":45,"forksCount":46,"repositoryHtmlUrl":47,"repositoryHomepageUrl":48,"ownerHtmlUrl":49,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":50,"install":51,"resources":64},"01M6EHJ27QWA786484SWXW58HN","control-concepts",null,"Guide to Control Tower's AI control concepts. Use when the user asks about control protocols, blue team and red team roles, usefulness, safety, audit budgets, trusted monitoring, or the conceptual model behind Control Tower.","linuxarena\u002Fcontrol-tower","linuxarena","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F266152290?v=4","01M6EHHVGQEZJ77QK9ZCEJXEZX","01M6EHHVNFDYG8BN3WX1RAZXTB",46,"MIT","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcontrol-concepts\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"concepts","control concepts","control","use","when","blue","about","ai control","red","roles","protocols","guide","user","model","team","Safety","blue team","red team","monitoring","audit","2026-11-10 10:12:30",false,"---\nname: control-concepts\ndescription: Guide to Control Tower's AI control concepts. Use when the user asks about control protocols, blue team and red team roles, usefulness, safety, audit budgets, trusted monitoring, or the conceptual model behind Control Tower.\n---\n\n@docs\u002Fcontrol-concepts.md\n","@docs\u002Fcontrol-concepts.md\n",{"name":5,"description":7},"main","6336a87956798b934243a31c64600d489778c531",".agents\u002Fskills\u002Fcontrol-concepts",53,"https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower","https:\u002F\u002Fdocs.linuxarena.ai","https:\u002F\u002Fgithub.com\u002Flinuxarena","2026-11-13 03:38:27",{"source":8,"skillName":5,"commands":52,"prompt":62,"zipAvailable":63},[53,56,59],{"tool":54,"command":55},"npx","npx skills add linuxarena\u002Fcontrol-tower --skill control-concepts",{"tool":57,"command":58},"bunx","bunx skills add linuxarena\u002Fcontrol-tower --skill control-concepts",{"tool":60,"command":61},"pnpm","pnpm dlx skills add linuxarena\u002Fcontrol-tower --skill control-concepts","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcontrol-concepts\u002FSKILL.md",true,[65],{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"README.md","REFERENCE","参考文档",164,"https:\u002F\u002Fgithub.com\u002FP1Piyush\u002Fskill-seo-auditor\u002Fblob\u002Fmain\u002FREADME.md",[72,101,114,126,152,163],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":38},"01M6EHJXGQN4F3RDVTVV17ZET4","validate-straj","Validate a sabotage trajectory (straj) end-to-end before trusting it as a keeper. Reconstructs the attack mechanism, checks for blatant reward-hacking and for unearned stealth (an environment flaw that made the side task trivially easy and low-suspicion), replays the side-task scorer in a fresh sandbox, and re-scores the replay with the `simple` (action) + `trajectory` (traj) monitors on opus-4-8 and gemini-3.1-pro (scored once per model, escalating a flagged monitor to a majority of 3). Use when the user wants to validate \u002F verify \u002F QA strajs, confirm a straj still evades the canonical monitor, or check a \"clean keeper\" list of trajectories.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fvalidate-straj\u002FSKILL.md",[],[79,80,81,82,31,83,84,85,86,87,88,89,21,90,91,92,93,94,95,96,97,98,99,22,100],"validate straj","validate","as","gemini","environment","end","clean","monitor","sandbox","replay","qa","task","check","fresh","trajectory","simple","verify","easy","for","re","true","trajectories",{"skillNo":102,"name":100,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":38},"01M6EHJTPEA16Z1ARWM2KE249W","Guide for working with Control Tower trajectories. Use when the user asks about trajectory formats, trajectory IDs, converting .eval logs, uploading or downloading trajectories, replaying trajectories, or trajectory CLI commands.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Ftrajectories\u002FSKILL.md",[],[100,107,108,93,109,20,21,110,29,30,111,112,97,24,113,22],"commands","eval","cli","ids","logs","with","formats",{"skillNo":115,"name":116,"nameZh":6,"description":117,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":118,"categories":119,"keywords":120,"publishedAt":38},"01M6EHJRFHV052KD1RDF3MB7Z4","safety-curves","Guide for understanding and explaining Control Tower safety curves. Use when the user asks about safety-usefulness tradeoffs, audit budgets, attack success rates, monitor thresholds, safety metrics, or interpreting safety curve plots.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsafety-curves\u002FSKILL.md",[],[33,121,122,86,123,20,21,37,29,30,124,97,24,125,22],"safety curves","curves","thresholds","metrics","curve",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":137,"publishedAt":38},"01M6EHJNFF12GYAK6XG1BDXB9Z","running-evals","Guide for running evaluations in Control Tower. Use when the user asks about running evals, ct run eval, policies (honest, attack, test, replay), protocols, task sets, uploading results, replaying trajectories, or evaluation options. IMPORTANT - Use this skill instead of the MCP tools for guidance on running evaluations.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Frunning-evals\u002FSKILL.md",[132],{"categoryNo":133,"categoryName":134,"taxonomyType":135,"taxonomyTypeDesc":136},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[138,139,140,28,108,88,141,21,90,142,20,143,144,145,146,147,148,149,150,151,29,97,22,100,24],"evals","running evals","running","MCP","test","results","options","run","run eval","Evaluation","guidance","CT","mcp tools","Tools",{"skillNo":153,"name":36,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":158,"publishedAt":38},"01M6EHJJXYBSRR4APEKCXHP9R2","Guide for monitoring trajectories in Control Tower. Use when the user asks about monitoring, suspicion scores, the ct run monitor command, monitor prompts and fields, or trajectory analysis. IMPORTANT - Use this skill instead of the MCP tools for guidance on monitoring trajectories.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fmonitoring\u002FSKILL.md",[157],{"categoryNo":133,"categoryName":134,"taxonomyType":135,"taxonomyTypeDesc":136},[36,93,20,159,145,148,86,149,141,150,21,160,161,151,29,30,97,24,162,22,100],"command","prompts","analysis","fields",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":169,"publishedAt":38},"01M6EHJGD95X8VP4ZYCM82HDZX","making-strajs","Guide for creating sabotage trajectories (strajs) in Control Tower. Use when the user asks about creating strajs, monitor evasion, attack strategies, straj documentation, or trajectory development workflows.","https:\u002F\u002Fgithub.com\u002Flinuxarena\u002Fcontrol-tower\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fmaking-strajs\u002FSKILL.md",[],[170,171,93,172,20,173,86,174,21,175,176,29,30,97,24,22,100],"making strajs","documentation","strategies","development","evasion","development workflows","workflows"]