[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EJFEAQRG489TA9997FMBKZ":3,"skill-same:01M7EJFEAQRG489TA9997FMBKZ":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},"01M7EJFEAQRG489TA9997FMBKZ","debugging",null,"Use when something is broken and you do not yet know why: a bug, a failing test, behaviour nobody expected, a test that fails only sometimes, or code that worked before. Triggers: bug, test failure, unexpected behaviour, flaky, it worked before, root cause, why does this fail, proc to pada. Not for restructuring working code (refactoring).","qsmark-labs\u002Fqsm-marketplace","qsmark-labs","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F348383745?v=4","01M7EJDK5MKZ67A7JBGC3ZB1RP","QSM-LABS","01M7EJDK6649PCGA2ENGFMFA7E",0,"GPL-3.0","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-essentials\u002Fskills\u002Fdebugging\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: debugging\ndescription: \"Use when something is broken and you do not yet know why: a bug, a failing test, behaviour nobody expected, a test that fails only sometimes, or code that worked before. Triggers: bug, test failure, unexpected behaviour, flaky, it worked before, root cause, why does this fail, proc to pada. Not for restructuring working code (refactoring).\"\n---\n\n# Debugging\n\nA fix made before the cause is known is a guess, and a guess that turns the check green hides the\nfault: it returns as a different symptom nobody connects to this one. This skill works in steps:\nreproduce, find the cause, make the one fix for it, then say what was wrong, how it was found and\nwhy the fix holds. It serves the person who must trust the fix without redoing the search. It does\nnot clean up around the fix (refactoring) and does not choose the regression test\n(software-testing).\n\n## Rules\n\n- Reproduce it first, as one command you can rerun, and run it yourself before any edit. Without\n  it you fix the bug you imagined and cannot tell when you are done. Cannot reproduce: say so and\n  ask for input, version and steps.\n- The same command is the proof after the fix. A different command is a different test, so a green\n  result from it says nothing about this bug.\n- Trace backwards from the bad value to the first place it was right and left wrong. With several\n  hops between symptom and origin, halve the path: look at the value at the middle hop, and the\n  half that is already wrong holds the cause. Fix there. A null check or try block at the symptom\n  leaves every other caller broken. Find callers with the code index's impact query (the query\n  that lists what depends on a symbol) before the edit.\n- A fallback on the path (`|| 0`, `?? \"\"`, an empty catch, a default) is where the trail goes\n  quiet, because it turns a wrong input into a plausible output. Make it throw for the run and the\n  first wrong value names itself. Whether it stays is a decision after the cause is known.\n- Change one variable per run, because two changes and one result prove nothing.\n- Name the cause in one sentence with a \"because\", and prove it by changing the one input it names:\n  the symptom must move. Drop a wrong hypothesis, never stack a second guess on it.\n- Something that looks wrong but does not move the symptom when changed is ruled out. Report it,\n  do not fix it, because a fix for it is a second change nobody asked for.\n- A flaky failure gets a rate: run the command in a loop (20 runs) and diff one pass against one\n  fail (order, timing, leftover state, machine). Retry, skip and weaken are never the fix.\n- It worked before: `git bisect run \u003Cthe reproduction>` between the last good commit and now.\n  A handful of runs finds the commit however large the range.\n- Set a budget (three hypotheses) before you start. When it runs out, stop and report what you\n  ruled out and what you would need. When three fixes in a row each expose a new problem, stop: it\n  is a design problem and the user decides.\n- Fix with one change for the cause and nothing \"while I am here\", because a reviewer cannot\n  separate a fix from a refactor. Rerun the reproduction, then the project's tests once.\n- Remove your instrumentation. Ask software-testing for the cheapest check that would have caught\n  this, because the same defect should fail loudly next time.\n- Record the cause in the commit message. A trap that will catch someone again goes into the\n  instructions; a defect you found elsewhere, or a smell, is a task in the task tool, not part of\n  this fix. Questions about logs, config or how it starts are run-and-verify.\n\n## Red flags\n\n- When you are about to edit and have not seen the failure yourself, reproduce it, because the\n  fix targets an imagined bug.\n- When the symptom went quiet after a guard or a retry, find what produced the bad value, because\n  the guard only moved the failure.\n- When the check is green and you cannot name the causing line, keep going, because the cause is\n  still in the code.\n- When the diff touches more than the line the cause names, split the rest out, because it is not\n  the fix.\n\n## Say it in the reply\n\nThree parts. What: the cause in one sentence with its \"because\". How: the command that reproduces\nit and the hops that led to the cause, with what was ruled out. Why: what the same command showed\nafter the fix and why the fix holds. When the cause is not proved yet, or someone else takes the\nwork over: each known fact with where it comes from (a run, a log, a file, a person's claim), kept\napart from what is only suspected, how the cause will be proved before anything is fixed, and what\nblocks you and what you need from whom, so the reader acts without asking back. When the budget\nran out: what you ruled out and what you would try next, never a fix without its cause.\n","# Debugging\n\nA fix made before the cause is known is a guess, and a guess that turns the check green hides the\nfault: it returns as a different symptom nobody connects to this one. This skill works in steps:\nreproduce, find the cause, make the one fix for it, then say what was wrong, how it was found and\nwhy the fix holds. It serves the person who must trust the fix without redoing the search. It does\nnot clean up around the fix (refactoring) and does not choose the regression test\n(software-testing).\n\n## Rules\n\n- Reproduce it first, as one command you can rerun, and run it yourself before any edit. Without\n  it you fix the bug you imagined and cannot tell when you are done. Cannot reproduce: say so and\n  ask for input, version and steps.\n- The same command is the proof after the fix. A different command is a different test, so a green\n  result from it says nothing about this bug.\n- Trace backwards from the bad value to the first place it was right and left wrong. With several\n  hops between symptom and origin, halve the path: look at the value at the middle hop, and the\n  half that is already wrong holds the cause. Fix there. A null check or try block at the symptom\n  leaves every other caller broken. Find callers with the code index's impact query (the query\n  that lists what depends on a symbol) before the edit.\n- A fallback on the path (`|| 0`, `?? \"\"`, an empty catch, a default) is where the trail goes\n  quiet, because it turns a wrong input into a plausible output. Make it throw for the run and the\n  first wrong value names itself. Whether it stays is a decision after the cause is known.\n- Change one variable per run, because two changes and one result prove nothing.\n- Name the cause in one sentence with a \"because\", and prove it by changing the one input it names:\n  the symptom must move. Drop a wrong hypothesis, never stack a second guess on it.\n- Something that looks wrong but does not move the symptom when changed is ruled out. Report it,\n  do not fix it, because a fix for it is a second change nobody asked for.\n- A flaky failure gets a rate: run the command in a loop (20 runs) and diff one pass against one\n  fail (order, timing, leftover state, machine). Retry, skip and weaken are never the fix.\n- It worked before: `git bisect run \u003Cthe reproduction>` between the last good commit and now.\n  A handful of runs finds the commit however large the range.\n- Set a budget (three hypotheses) before you start. When it runs out, stop and report what you\n  ruled out and what you would need. When three fixes in a row each expose a new problem, stop: it\n  is a design problem and the user decides.\n- Fix with one change for the cause and nothing \"while I am here\", because a reviewer cannot\n  separate a fix from a refactor. Rerun the reproduction, then the project's tests once.\n- Remove your instrumentation. Ask software-testing for the cheapest check that would have caught\n  this, because the same defect should fail loudly next time.\n- Record the cause in the commit message. A trap that will catch someone again goes into the\n  instructions; a defect you found elsewhere, or a smell, is a task in the task tool, not part of\n  this fix. Questions about logs, config or how it starts are run-and-verify.\n\n## Red flags\n\n- When you are about to edit and have not seen the failure yourself, reproduce it, because the\n  fix targets an imagined bug.\n- When the symptom went quiet after a guard or a retry, find what produced the bad value, because\n  the guard only moved the failure.\n- When the check is green and you cannot name the causing line, keep going, because the cause is\n  still in the code.\n- When the diff touches more than the line the cause names, split the rest out, because it is not\n  the fix.\n\n## Say it in the reply\n\nThree parts. What: the cause in one sentence with its \"because\". How: the command that reproduces\nit and the hops that led to the cause, with what was ruled out. Why: what the same command showed\nafter the fix and why the fix holds. When the cause is not proved yet, or someone else takes the\nwork over: each known fact with where it comes from (a run, a log, a file, a person's claim), kept\napart from what is only suspected, how the cause will be proved before anything is fixed, and what\nblocks you and what you need from whom, so the reader acts without asking back. When the budget\nran out: what you ruled out and what you would try next, never a fix without its cause.\n",{"name":5,"description":7},"main","c385fc50aadaee7296445e049153ce0560a9c80c","plugins\u002Fqsm-essentials\u002Fskills\u002Fdebugging","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace","https:\u002F\u002Fgithub.com\u002Fqsmark-labs","2026-11-13 18:31:51",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add qsmark-labs\u002Fqsm-marketplace --skill debugging",{"tool":36,"command":37},"bunx","bunx skills add qsmark-labs\u002Fqsm-marketplace --skill debugging",{"tool":39,"command":40},"pnpm","pnpm dlx skills add qsmark-labs\u002Fqsm-marketplace --skill debugging","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-essentials\u002Fskills\u002Fdebugging\u002FSKILL.md",true,[],[45,52,59,66,73,80],{"skillNo":46,"name":47,"nameZh":6,"description":48,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":19},"01M7EJWPJ0B4E1P148MWQ3GMR8","worktrunk","Use when working with git worktrees or wt (worktrunk): creating, switching, merging or removing a branch workspace, configuring wt (project or personal config, workflow hooks around switch, commit and merge, commit-message generation, hook approvals), or a wt command that failed. Triggers: make a worktree for this branch, wt merge, wt remove, configure wt, .config\u002Fwt.toml, run tests before merge, install deps when a worktree is created, wt hook, wt switch does not cd, nastav wt, wt nefunguje.","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-worktrunk\u002Fskills\u002Fworktrunk\u002FSKILL.md",[],[],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M7EJWD4F683TF21N89JGWZB2","web-ladder","Use when a task needs something the internet holds: look something up, search the web, find which page answers a question, read a page, check a library's current docs, a page that needs a browser or shows a captcha. Runs one research loop: searxng search, keywords, cases, search again up to three times, Context7 for a library, Tavily when the keywords will not come, one page in a browser, the person on a wall. Triggers: look this up, search the web, find out online, read this page, check the docs, the page is blocked, captcha, najdi na webu, otevri tu stranku. A multi-round question goes to the qsm-web:web-research agent; one tool's commands are its own skill (searxng, browserbox, opensrc).","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-web\u002Fskills\u002Fweb-ladder\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7EJW9G0K6AS5Z4H0B0K6NDB","searxng","Use when running a web search through the person's own SearXNG instance with qsm-web searxng: choosing a search profile or engines, the instance, time range or language, reading its JSON, a suspended engine. Triggers: search with searxng, qsm-web searxng, search profile, niche search, academic search, IT search, vyhledavaci profil, searxng is down, which engines. The research loop around it is the web-ladder skill.","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-web\u002Fskills\u002Fsearxng\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"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":72,"publishedAt":19},"01M7EJW40RD3DRBQVVCW6K7NMG","opensrc","Use when the source code of an open-source library or repository is needed on disk to read: the docs are silent, the behaviour must be checked in code, the version the project pins. Runs qsm-web opensrc get with a package spec or a repository address. Triggers: get the source of this package, clone this library, read the library's code, qsm-web opensrc, stahni zdrojak knihovny. Library docs are the Context7 skill; the research loop is the web-ladder skill.","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-web\u002Fskills\u002Fopensrc\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7EJVZWD6J7W91G69GMBWBQS","browserbox","Use when one page needs a real browser through qsm-web browserbox: content built by script, a page that blocks plain fetching, a captcha to solve or hand to the person, a scripted sequence of clicks, a login the person is present for. Triggers: open it in a browser, the page is empty without JavaScript, captcha, take over the browser, run these steps on the page, log in to this site, browserbox. When to reach for a browser at all is the web-ladder skill.","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-web\u002Fskills\u002Fbrowserbox\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7EJVSZC1E8ZGVE8KXVFPR9S","usage","Use when the user wants a review of how they use Claude Code, or setup hands over after a first install: review my Claude Code usage, audit my setup, what does my setup cost, where did my tokens go, clean up my Claude Code, usage report, zkontroluj moje pouzivani. One walkthrough of four stages (machine, work, habits, extensions), questions after each, then one plan. A question about one topic (which hooks are slow) goes to that topic's skill; this session's tokens alone is `\u002Fqsm-usage`.","https:\u002F\u002Fgithub.com\u002Fqsmark-labs\u002Fqsm-marketplace\u002Fblob\u002Fmain\u002Fplugins\u002Fqsm-usage\u002Fskills\u002Fusage\u002FSKILL.md",[],[]]