[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EP6XB5TAM6PSXGDRSAGT2F":3,"skill-same:01M7EP6XB5TAM6PSXGDRSAGT2F":44},{"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":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M7EP6XB5TAM6PSXGDRSAGT2F","unexercised-branch",null,"A check that never met the case it guards is not yet a check. Three faces from one day (2026-10-11): a hook that rewrites text touched a record because \"lines this commit adds\" looked like \"new writing\"; a fresh agent that had not seen the work found twelve failures the specs passed; a shelf sealer's inverted guard hid behind a fixture whose failing check failed silently, so the refusal branch never ran. Use when a spec passes on its first run, when a guard refuses (seal, publish, promote) and no test makes it refuse, when a hook or tool rewrites text in commits, on \"is this branch ever reached\", \"did the test fail for the right reason\", \"fresh eyes\", \"mutation-test the spec\", or \"is this a record\". Electable. PROVISIONAL.","inputloop\u002Fgarden","inputloop","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F339507215?v=4","01M7ENYT15EMVBAV30YXPQDYE2","Inputloop","01M7ENYT1SM9WM5PSMQDC9PE2G",0,"GPL-3.0","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002FSOFT-SKILLS\u002Funexercised-branch\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: unexercised-branch\ndescription: >-\n  A check that never met the case it guards is not yet a check. Three faces from one day\n  (2026-10-11): a hook that rewrites text touched a record because \"lines this commit adds\" looked\n  like \"new writing\"; a fresh agent that had not seen the work found twelve failures the specs\n  passed; a shelf sealer's inverted guard hid behind a fixture whose failing check failed silently,\n  so the refusal branch never ran. Use when a spec passes on its first run, when a guard refuses\n  (seal, publish, promote) and no test makes it refuse, when a hook or tool rewrites text in commits,\n  on \"is this branch ever reached\", \"did the test fail for the right reason\", \"fresh eyes\",\n  \"mutation-test the spec\", or \"is this a record\". Electable. PROVISIONAL.\n---\n\n# unexercised-branch: the check that never met its case\n\nLanded 2026-10-11 at his ask to bring the day's learnings into SOFT-SKILLS. The receipts are in\n`\u002Ff\u002Fclaude\u002Fc4\u002FMEMORY.md` § 2026-10-11 (late afternoon) and\n`\u002Ff\u002Fclaude\u002Fc4\u002Fh\u002FENHANCE-2026-10-11-learnings.md`, learnings 4-6.\n\n## The idea in one paragraph\n\nA check proves what it has been seen to do. A guard whose refusal no test reaches, a spec that\npassed the first time it ran, and a rewrite rule that has never been shown an old text in a new\nplace are each a branch nobody has walked. The fault lives there because nobody looked there. The\ncure is cheap: make the branch run, once, on purpose, and keep the case.\n\n## The three moves\n\n1. **Records are not new writing.** A tool that rewrites text (spelling, paths, headers) acts only\n   on writing, never on records. \"Lines this commit adds\" is not the test: an archive copy, a\n   restored file, a keep, a quoted transcript are new to git and old in words. Before rewriting,\n   ask \"is this a record?\"; if the tool cannot tell, it leaves the file alone and says so.\n   OBSERVED: the pa1th hook's first live run changed an archive keep; restored byte for byte\n   (bb4417c3), c4 commit 433ec74.\n2. **A spec green on its first run gets mutation-tested.** Run it against the old code (or a copy\n   with the fix taken out). If it still passes, it does not test the fix. Then hand the work to a\n   fresh agent that has not seen it, with the specs but not the story, and ask it to break the\n   work. Every repro it finds becomes a case. OBSERVED: twelve failures, c4 commit dcb69ad.\n3. **A refusing guard is proven by a planted failure that speaks.** For every guard that refuses,\n   one case plants a failure and expects the refusal, and the planted failure must print a named\n   line, so a silent fixture cannot stand in for a refusal. OBSERVED: `bin\u002Fsh4lf` sealed past a\n   failing check; its test's check failed with no output, so the branch never ran. sh4lf's spec\n   now holds 17 cases.\n\n## Where it does not apply\n\n- A check with no branch (a pure count, such as `bin\u002Fvo1ce`, which gives no verdict).\n- A one-off probe that is thrown away. Once it is kept, this applies.\n\n## Falsifiers\n\n- Move 1: a rewrite tool, given only \"lines this commit adds\", that never touched a record over a\n  month of commits.\n- Move 2: a first-run-green spec that, mutation-tested, always failed against the old code.\n- Move 3: a refusing guard with no planted-failure case that was never found fail-open.\n\n## Neighbours\n\ngrand-test-manifest (every promise checked on the real work and on a planted copy) is the general\nform of move 3; depth-overlap asks whether two checks see one promise from different places. This\nsoft skill is the narrower, day-sized lesson; if grand-test-manifest already says it, fold this in\nand remove the folder.\n\nOff-switch: say \"no branch check on this\", or remove this folder.\n\n---\n\nCopyright 2026 Corwin Brust \u003Ccorwin@bru.st> and Contributors; CC-BY-SA4: https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-sa\u002F4.0\u002F; Mentat (for Corwin; claude-opus-5-5)\n","# unexercised-branch: the check that never met its case\n\nLanded 2026-10-11 at his ask to bring the day's learnings into SOFT-SKILLS. The receipts are in\n`\u002Ff\u002Fclaude\u002Fc4\u002FMEMORY.md` § 2026-10-11 (late afternoon) and\n`\u002Ff\u002Fclaude\u002Fc4\u002Fh\u002FENHANCE-2026-10-11-learnings.md`, learnings 4-6.\n\n## The idea in one paragraph\n\nA check proves what it has been seen to do. A guard whose refusal no test reaches, a spec that\npassed the first time it ran, and a rewrite rule that has never been shown an old text in a new\nplace are each a branch nobody has walked. The fault lives there because nobody looked there. The\ncure is cheap: make the branch run, once, on purpose, and keep the case.\n\n## The three moves\n\n1. **Records are not new writing.** A tool that rewrites text (spelling, paths, headers) acts only\n   on writing, never on records. \"Lines this commit adds\" is not the test: an archive copy, a\n   restored file, a keep, a quoted transcript are new to git and old in words. Before rewriting,\n   ask \"is this a record?\"; if the tool cannot tell, it leaves the file alone and says so.\n   OBSERVED: the pa1th hook's first live run changed an archive keep; restored byte for byte\n   (bb4417c3), c4 commit 433ec74.\n2. **A spec green on its first run gets mutation-tested.** Run it against the old code (or a copy\n   with the fix taken out). If it still passes, it does not test the fix. Then hand the work to a\n   fresh agent that has not seen it, with the specs but not the story, and ask it to break the\n   work. Every repro it finds becomes a case. OBSERVED: twelve failures, c4 commit dcb69ad.\n3. **A refusing guard is proven by a planted failure that speaks.** For every guard that refuses,\n   one case plants a failure and expects the refusal, and the planted failure must print a named\n   line, so a silent fixture cannot stand in for a refusal. OBSERVED: `bin\u002Fsh4lf` sealed past a\n   failing check; its test's check failed with no output, so the branch never ran. sh4lf's spec\n   now holds 17 cases.\n\n## Where it does not apply\n\n- A check with no branch (a pure count, such as `bin\u002Fvo1ce`, which gives no verdict).\n- A one-off probe that is thrown away. Once it is kept, this applies.\n\n## Falsifiers\n\n- Move 1: a rewrite tool, given only \"lines this commit adds\", that never touched a record over a\n  month of commits.\n- Move 2: a first-run-green spec that, mutation-tested, always failed against the old code.\n- Move 3: a refusing guard with no planted-failure case that was never found fail-open.\n\n## Neighbours\n\ngrand-test-manifest (every promise checked on the real work and on a planted copy) is the general\nform of move 3; depth-overlap asks whether two checks see one promise from different places. This\nsoft skill is the narrower, day-sized lesson; if grand-test-manifest already says it, fold this in\nand remove the folder.\n\nOff-switch: say \"no branch check on this\", or remove this folder.\n\n---\n\nCopyright 2026 Corwin Brust \u003Ccorwin@bru.st> and Contributors; CC-BY-SA4: https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby-sa\u002F4.0\u002F; Mentat (for Corwin; claude-opus-5-5)\n",{"name":5,"description":7},"main","0b0a0f665f67605c3bc1d43500eea26ef3079a97","SOFT-SKILLS\u002Funexercised-branch","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden","https:\u002F\u002Fgithub.com\u002Finputloop","2026-10-14 18:29:39",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add inputloop\u002Fgarden --skill unexercised-branch",{"tool":36,"command":37},"bunx","bunx skills add inputloop\u002Fgarden --skill unexercised-branch",{"tool":39,"command":40},"pnpm","pnpm dlx skills add inputloop\u002Fgarden --skill unexercised-branch","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002FSOFT-SKILLS\u002Funexercised-branch\u002FSKILL.md",true,[],[45,51,57,64,71,78],{"skillNo":46,"name":47,"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":48,"categories":49,"keywords":50,"publishedAt":19},"01M7EPTX6X52ZPGVQZHKVJ7FAZ","alpha","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flane-d\u002Fbr1in\u002Fours\u002Fskills\u002Falpha\u002FSKILL.md",[],[],{"skillNo":52,"name":53,"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":54,"categories":55,"keywords":56,"publishedAt":19},"01M7EPTWCTS80QHFQ2ND3A1CW1","gamma","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flane-d\u002Fbr1in\u002Fours\u002FSOFT-SKILLS\u002Fgamma\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M7EPTV3K3SCTRP0DR6BMMKR0","quiet","A fixture skill that declares no preferences, so skill.quiet.* has no reader.","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flane-a\u002Fdeclared\u002Fsoft\u002Fquiet\u002FSKILL.md",[],[],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7EPTSG0VE67F6PFAC0XJDQ4","demo","A fixture skill for tests\u002Fprefs-declared-spec.rb; it declares two preferences inline.","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flane-a\u002Fdeclared\u002Fsoft\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7EPTP0PM5WPTFBHPXBCFM5C","concise","Synthesized by polygram from memory tag 'concise'. explicitly flagged. Review and tighten this description before relying on it for triggering.","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fconcise\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7EPTAETQ62DJA8ZFEEBBZF0","writing-forms","Which writing form a dated deliverable takes in the shop, and the five techniques every form shares. Use whenever someone is about to write a dated deliverable — \"which form is this?\", \"write it up\", \"land a note \u002F card \u002F model \u002F lens\", \"what shape does a … take\", \"is there a form for this\", or when a new PREFIX-date-slug filename is about to be minted. Carries the rule that a deliverable picks ONE form and says so in its filename, points at the register c\u002Fwf (columns d\u002Fc\u002Fwf) as the source of truth, names the most-used forms, the five cross-cutting techniques (two piles · cite at the act · a falsifier on every concern · the pair rule · his words), how a new form is born, and the fence on notes other sessions wrote. Not rolls (report kinds and their guards), not reformat-prompt (the REFORMAT pair), not run-card (the RUNCARD), not skill-creator (skills). Owns the FORM, not the content.","https:\u002F\u002Fgithub.com\u002Finputloop\u002Fgarden\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-forms\u002FSKILL.md",[],[]]