[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DGYA15VD356QPCVE8KCSZV":3,"skill-same:01M7DGYA15VD356QPCVE8KCSZV":45},{"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":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M7DGYA15VD356QPCVE8KCSZV","prove-it",null,"Separates what was checked from what was only written before a coding task is reported as done. Use when finishing a code change, bug fix, or refactor, and before saying anything is fixed, working, passing, or complete. Pairs each claim with the command and output that backs it, and lists what was not verified.","Xnadir\u002Fvetted","Xnadir","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F36660597?v=4","01M7DGX6VS1V5A30PQHWDZ4ECE","Nadir Ali","01M7DGX6WAECNN13JBVANGT6BG",5,"MIT","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Fprove-it\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: prove-it\ndescription: Separates what was checked from what was only written before a coding task is reported as done. Use when finishing a code change, bug fix, or refactor, and before saying anything is fixed, working, passing, or complete. Pairs each claim with the command and output that backs it, and lists what was not verified.\nlicense: MIT\n---\n\n# Prove it\n\n\"Done\" is a claim about the world, not about what you typed. People act on your\nfinal message: they merge, deploy, and close tickets. An honest \"not verified\"\ncosts them a minute. A false \"fixed\" can cost them a day, and it costs you\ntheir trust in every later report.\n\n## Before you report\n\n1. List the claims your final message will make: \"the bug is fixed\", \"tests\n   pass\", \"the endpoint now returns 404 for unknown users\".\n2. For each claim, run the strongest check available right now:\n   - a test that exercises the change (narrowest target first, then the suite)\n   - the build, type-checker, or linter for the files you touched\n   - the program or the original reproduction, with the output observed\n   - reading code backs claims about structure (\"the function is now pure\"),\n     never claims about behavior (\"it works\")\n3. Checks must run after your last edit. A run from before the final edit does\n   not cover it.\n4. If a check fails, you are not done. Fix it, or report the failure plainly.\n\nNever make a check pass by weakening it: no skipped or deleted tests, loosened\nassertions, swallowed exceptions, or `|| true`. If you believe a test is wrong,\nsay why and ask.\n\n## How to report\n\nEnd with a short evidence block. Quote real output such as counts, exit codes,\nand error lines, rather than paraphrasing it as \"all good\".\n\n```\nVerified\n- `npm test -- auth.spec.ts` → 14 passed (run after the final edit)\n- Original repro `curl -i localhost:3000\u002Fusers\u002F999` → was 500, now 404\nNot verified\n- Full e2e suite (needs Docker, not available here)\n- Windows path handling\n```\n\nWording rules:\n\n- State behavior as fact only when a line under \"Verified\" backs it.\n  Everything else is \"should\" and goes under \"Not verified\" with the reason.\n- If you could not run anything at all (no shell, no runtime, missing\n  dependencies), say so in the first line of the report, not at the end.\n- Give the user the exact command to verify it themselves when you could not.\n\n## Scale it to the task\n\n- A one-line change with an obvious check needs one line of evidence.\n- A question with no change made: this skill does not apply.\n- A long task: verify as you go; the final report covers the final state only.\n","# Prove it\n\n\"Done\" is a claim about the world, not about what you typed. People act on your\nfinal message: they merge, deploy, and close tickets. An honest \"not verified\"\ncosts them a minute. A false \"fixed\" can cost them a day, and it costs you\ntheir trust in every later report.\n\n## Before you report\n\n1. List the claims your final message will make: \"the bug is fixed\", \"tests\n   pass\", \"the endpoint now returns 404 for unknown users\".\n2. For each claim, run the strongest check available right now:\n   - a test that exercises the change (narrowest target first, then the suite)\n   - the build, type-checker, or linter for the files you touched\n   - the program or the original reproduction, with the output observed\n   - reading code backs claims about structure (\"the function is now pure\"),\n     never claims about behavior (\"it works\")\n3. Checks must run after your last edit. A run from before the final edit does\n   not cover it.\n4. If a check fails, you are not done. Fix it, or report the failure plainly.\n\nNever make a check pass by weakening it: no skipped or deleted tests, loosened\nassertions, swallowed exceptions, or `|| true`. If you believe a test is wrong,\nsay why and ask.\n\n## How to report\n\nEnd with a short evidence block. Quote real output such as counts, exit codes,\nand error lines, rather than paraphrasing it as \"all good\".\n\n```\nVerified\n- `npm test -- auth.spec.ts` → 14 passed (run after the final edit)\n- Original repro `curl -i localhost:3000\u002Fusers\u002F999` → was 500, now 404\nNot verified\n- Full e2e suite (needs Docker, not available here)\n- Windows path handling\n```\n\nWording rules:\n\n- State behavior as fact only when a line under \"Verified\" backs it.\n  Everything else is \"should\" and goes under \"Not verified\" with the reason.\n- If you could not run anything at all (no shell, no runtime, missing\n  dependencies), say so in the first line of the report, not at the end.\n- Give the user the exact command to verify it themselves when you could not.\n\n## Scale it to the task\n\n- A one-line change with an obvious check needs one line of evidence.\n- A question with no change made: this skill does not apply.\n- A long task: verify as you go; the final report covers the final state only.\n",{"name":5,"license":15,"description":7},"main","460479d90ead1f6731b5ff91d61101a65879299a","skills\u002Fprove-it",8,"https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted","https:\u002F\u002Fgithub.com\u002FXnadir","2026-10-30 15:30:58",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add Xnadir\u002Fvetted --skill prove-it",{"tool":37,"command":38},"bunx","bunx skills add Xnadir\u002Fvetted --skill prove-it",{"tool":40,"command":41},"pnpm","pnpm dlx skills add Xnadir\u002Fvetted --skill prove-it","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Fprove-it\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"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},"01M7DGYD618W0F8XR4TJGEKC6M","root-cause","Finds and fixes the actual cause of a bug instead of patching the symptom at the place it was reported. Use when debugging an error, a failing test, a crash, or wrong output, and when a bug report names one caller, page, or input. Traces the failure back to where the bad state originates and fixes it once there.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Froot-cause\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},"01M7DGY5CP1C99S0GESEYNCT1C","handoff","Writes a handoff note that lets a fresh agent session or a teammate resume the current task without re-reading the conversation. Use when the user asks for a handoff, summary for later, or context dump, or before ending a long session, switching machines, or clearing context mid-task.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Fhandoff\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},"01M7DGY0TMQBGBMWSPK8BDTAJH","grill","Interviews the user about an under-specified feature, project, or plan before any code is written, one decision at a time with a recommended answer for each. Use when the user asks to be grilled, interviewed, or questioned about requirements, a plan, or a design before building.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Fgrill\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},"01M7DGXWE6VKYKG2QCBNWCPVZ5","bug-hunt-review","Reviews code or a diff for real defects, meaning inputs or states that produce wrong results, crashes, data loss, or security holes, and reports each with a concrete failure scenario. Use when asked to review code, a pull request, a diff, or a change before merge. Leaves out style nitpicks unless asked.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fskills\u002Fbug-hunt-review\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},"01M7DGXQKRA47TFV2YJCBNDDFJ","surgical","Keeps code changes to the smallest diff that fully solves the request. Use when editing existing code to fix a bug, add a feature, or respond to review, especially in files with unrelated code nearby. Prevents drive-by refactors, reformatting, renames, and \"while I'm here\" cleanups that make diffs hard to review.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fretired\u002Fskills\u002Fsurgical\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},"01M7DGXJSBPVBC9N0F1NXYNEY9","stdlib-first","Solves small programming problems with the language's standard library, platform features, or already-installed packages before adding a new dependency. Use when writing code that could pull in a package (dates, deep clone, UUIDs, HTTP, arg parsing, formatting, validation), or when the user asks which library to use.","https:\u002F\u002Fgithub.com\u002FXnadir\u002Fvetted\u002Fblob\u002Fmain\u002Fretired\u002Fskills\u002Fstdlib-first\u002FSKILL.md",[],[]]