[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DHQGEMHHGRMZ4R46SA59Z7":3,"skill-same:01M6DHQGEMHHGRMZ4R46SA59Z7":66},{"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":36,"linkOnly":37,"markdownContent":38,"markdownBody":39,"frontMatter":40,"refName":43,"commitSha":44,"skillDirectory":45,"forksCount":14,"repositoryHtmlUrl":46,"repositoryHomepageUrl":47,"ownerHtmlUrl":48,"ownerBlogUrl":49,"ownerDescription":50,"updatedAt":51,"install":52,"resources":65},"01M6DHQGEMHHGRMZ4R46SA59Z7","opsward",null,"Run opsward to assess and improve this project's AI agent setup. Use when the user says 'opsward', 'check my setup', 'improve my AI config', or wants a full audit and remediation.","thorwhalen\u002Fopsward","thorwhalen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1906276?v=4","01M6DHNW6H23WSXQ1D6MJCFC91","Thor Whalen","01M6DHNWD2G3GJ64J30A1FW0XP",0,"MIT","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35],"use","when","agent setup","my","assess","full","project","ai config","ai agent","user","check","improve","remediation","audit","run","setup","config","2026-11-10 10:12:30",false,"---\nname: \"run\"\ndescription: \"Run a single experiment iteration. Edit the target file, evaluate, keep or discard. Use when the user runs \u002Far:run or asks for one manual autoresearch iteration.\"\ncommand: \u002Far:run\n---\n\n# \u002Far:run — Single Experiment Iteration\n\nRun exactly ONE experiment iteration: review history, decide a change, edit, commit, evaluate.\n\n## Usage\n\n```\n\u002Far:run engineering\u002Fapi-speed              # Run one iteration\n\u002Far:run                                     # List experiments, let user pick\n```\n\n## What It Does\n\n### Step 1: Resolve experiment\n\nIf no experiment specified, run `python {skill_path}\u002Fscripts\u002Fsetup_experiment.py --list` and ask the user to pick.\n\n### Step 2: Load context\n\n```bash\n# Read experiment config\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fconfig.cfg\n\n# Read strategy and constraints\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fprogram.md\n\n# Read experiment history\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fresults.tsv\n\n# Checkout the experiment branch\ngit checkout autoresearch\u002F{domain}\u002F{name}\n```\n\n### Step 3: Decide what to try\n\nReview results.tsv:\n- What changes were kept? What pattern do they share?\n- What was discarded? Avoid repeating those approaches.\n- What crashed? Understand why.\n- How many runs so far? (Escalate strategy accordingly)\n\n**Strategy escalation:**\n- Runs 1-5: Low-hanging fruit (obvious improvements)\n- Runs 6-15: Systematic exploration (vary one parameter)\n- Runs 16-30: Structural changes (algorithm swaps)\n- Runs 30+: Radical experiments (completely different approaches)\n\n### Step 4: Make ONE change\n\nEdit only the target file specified in config.cfg. Change one thing. Keep it simple.\n\n### Step 5: Commit and evaluate\n\n```bash\ngit add {target}\ngit commit -m \"experiment: {short description of what changed}\"\n\npython {skill_path}\u002Fscripts\u002Frun_experiment.py \\\n  --experiment {domain}\u002F{name} --single\n```\n\n### Step 6: Report result\n\nRead the script output. Tell the user:\n- **KEEP**: \"Improvement! {metric}: {value} ({delta} from previous best)\"\n- **DISCARD**: \"No improvement. {metric}: {value} vs best {best}. Reverted.\"\n- **CRASH**: \"Evaluation failed: {reason}. Reverted.\"\n\n### Step 7: Self-improvement check\n\nAfter every 10th experiment (check results.tsv line count), update the Strategy section of program.md with patterns learned.\n\n## Rules\n\n- ONE change per iteration. Don't change 5 things at once.\n- NEVER modify the evaluator (evaluate.py). It's ground truth.\n- Simplicity wins. Equal performance with simpler code is an improvement.\n- No new dependencies.\n","# \u002Far:run — Single Experiment Iteration\n\nRun exactly ONE experiment iteration: review history, decide a change, edit, commit, evaluate.\n\n## Usage\n\n```\n\u002Far:run engineering\u002Fapi-speed              # Run one iteration\n\u002Far:run                                     # List experiments, let user pick\n```\n\n## What It Does\n\n### Step 1: Resolve experiment\n\nIf no experiment specified, run `python {skill_path}\u002Fscripts\u002Fsetup_experiment.py --list` and ask the user to pick.\n\n### Step 2: Load context\n\n```bash\n# Read experiment config\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fconfig.cfg\n\n# Read strategy and constraints\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fprogram.md\n\n# Read experiment history\ncat .autoresearch\u002F{domain}\u002F{name}\u002Fresults.tsv\n\n# Checkout the experiment branch\ngit checkout autoresearch\u002F{domain}\u002F{name}\n```\n\n### Step 3: Decide what to try\n\nReview results.tsv:\n- What changes were kept? What pattern do they share?\n- What was discarded? Avoid repeating those approaches.\n- What crashed? Understand why.\n- How many runs so far? (Escalate strategy accordingly)\n\n**Strategy escalation:**\n- Runs 1-5: Low-hanging fruit (obvious improvements)\n- Runs 6-15: Systematic exploration (vary one parameter)\n- Runs 16-30: Structural changes (algorithm swaps)\n- Runs 30+: Radical experiments (completely different approaches)\n\n### Step 4: Make ONE change\n\nEdit only the target file specified in config.cfg. Change one thing. Keep it simple.\n\n### Step 5: Commit and evaluate\n\n```bash\ngit add {target}\ngit commit -m \"experiment: {short description of what changed}\"\n\npython {skill_path}\u002Fscripts\u002Frun_experiment.py \\\n  --experiment {domain}\u002F{name} --single\n```\n\n### Step 6: Report result\n\nRead the script output. Tell the user:\n- **KEEP**: \"Improvement! {metric}: {value} ({delta} from previous best)\"\n- **DISCARD**: \"No improvement. {metric}: {value} vs best {best}. Reverted.\"\n- **CRASH**: \"Evaluation failed: {reason}. Reverted.\"\n\n### Step 7: Self-improvement check\n\nAfter every 10th experiment (check results.tsv line count), update the Strategy section of program.md with patterns learned.\n\n## Rules\n\n- ONE change per iteration. Don't change 5 things at once.\n- NEVER modify the evaluator (evaluate.py). It's ground truth.\n- Simplicity wins. Equal performance with simpler code is an improvement.\n- No new dependencies.\n",{"name":33,"command":41,"description":42},"\u002Far:run","Run a single experiment iteration. Edit the target file, evaluate, keep or discard. Use when the user runs \u002Far:run or asks for one manual autoresearch iteration.","main","510dd2917e051cc953dc55f56fd6ef442aa01b38","opsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward","https:\u002F\u002Fthorwhalen.github.io\u002Fopsward\u002F","https:\u002F\u002Fgithub.com\u002Fthorwhalen","https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fthorwhalen\u002F","Machine Learning and AI Systems","2026-11-09 01:24:54",{"source":8,"skillName":33,"commands":53,"prompt":63,"zipAvailable":64},[54,57,60],{"tool":55,"command":56},"npx","npx skills add thorwhalen\u002Fopsward --skill run",{"tool":58,"command":59},"bunx","bunx skills add thorwhalen\u002Fopsward --skill run",{"tool":61,"command":62},"pnpm","pnpm dlx skills add thorwhalen\u002Fopsward --skill run","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward\u002FSKILL.md",true,[],[67,79,108,124,138,148],{"skillNo":68,"name":69,"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":70,"categories":71,"keywords":77,"publishedAt":36},"01M6DHQKJPZBWKFCJE4AER997W","my-skill","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fpython_project\u002F.claude\u002Fskills\u002Fmy-skill\u002FSKILL.md",[72],{"categoryNo":73,"categoryName":74,"taxonomyType":75,"taxonomyTypeDesc":76},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[22,78],"my skill",{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":86,"publishedAt":36},"01M6DHQCY2SHZ1ZEH5FQFT39WX","opsward-subagents","Decide which of a repo's skills should also run as subagents (in their own context), declare it in each skill's frontmatter, and generate or refresh the twin agent files with opsward. Use when a skill does bulky or read-heavy work whose output should stay out of the main context, when a skill needs a tool restriction or a cheaper model, when a hand-written agent duplicates a skill, when `opsward maintain` reports a `subagent_twin` finding, or when asked \"should this skill be a subagent\", \"make this skill usable as a subagent\", \"generate agents from skills\", or to sweep a repo for skill\u002Fagent pairs. Also usable as a subagent itself.","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward-subagents\u002FSKILL.md",[85],{"categoryNo":73,"categoryName":74,"taxonomyType":75,"taxonomyTypeDesc":76},[87,88,5,89,20,90,91,92,93,94,95,43,96,97,98,19,99,100,101,102,103,104,105,106,107],"subagents","opsward subagents","duplicates","for","which","true","opsward maintain","as","maintain","files","reports","generate","refresh","repo","decide","work","sweep","model","tool","subagent","Bash",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":36},"01M6DHQ9EGVVKAP608FD4B4FYA","opsward-maintain","Check documentation and AI setup for staleness, drift, or inconsistency. Use when the user asks to maintain, refresh, or update project docs and AI configuration.","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward-maintain\u002FSKILL.md",[],[5,93,95,29,115,116,20,90,117,118,25,119,120,121,122,19,28,99,123,34],"documentation","project docs","update","ai setup","check documentation","drift","configuration","update project docs","docs",{"skillNo":125,"name":126,"nameZh":6,"description":127,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":36},"01M6DHQ6F44PQADNEPRG7QAMPA","opsward-generate","Generate missing AI setup artifacts for this project. Use when the user asks to scaffold, create, or bootstrap CLAUDE.md, docs, skills, or agent configuration.","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward-generate\u002FSKILL.md",[],[98,5,131,132,133,20,90,118,134,25,121,19,28,135,136,34,137,123],"opsward generate","missing","create","agent configuration","artifacts","bootstrap","scaffold",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":144,"publishedAt":36},"01M6DHQ3YJMTQ155ZGHR76PP90","opsward-diagnose","Diagnose the health of this project's AI agent setup. Use when the user asks to check, audit, or score the project's CLAUDE.md, skills, docs, or agent configuration.","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002Fopsward\u002Fdata\u002Ftemplates\u002Fshared\u002Fopsward-diagnose\u002FSKILL.md",[],[145,5,146,29,147,32,34,27,123,28,20,21,134,25,121,19],"opsward diagnose","diagnose","health",{"skillNo":149,"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":150,"categories":151,"keywords":152,"publishedAt":36},"01M6DHQ1084K7TFC376MWMW3CQ","https:\u002F\u002Fgithub.com\u002Fthorwhalen\u002Fopsward\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fopsward\u002FSKILL.md",[],[5,25,20,21,22,23,24,26,19,28,29,30,31,32,33,34,35,27]]