[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NPWRZ4WCFBYTRA1NJ8H5K9":3,"skill-same:01M6NPWRZ4WCFBYTRA1NJ8H5K9":78},{"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":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":14,"repositoryHtmlUrl":55,"repositoryHomepageUrl":56,"ownerHtmlUrl":57,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":58,"install":59,"resources":72},"01M6NPWRZ4WCFBYTRA1NJ8H5K9","debugging",null,"Systematic debugging — reproduce, read the error, isolate, hypothesize, fix, verify, document. Use when tracking down a bug, unexpected behavior, or test failure — including phrases like \"this doesn't work\", \"why is this failing\", or \"help me debug\".","projectious-work\u002Fprocesskit","projectious-work","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F263889726?v=4","01M6N97TJJ118WQ35CHRV8RD0E","projectious.work","01M6N97TPWJPF0BMKYQ4PBY5W8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fengineering\u002Fdebugging\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"behavior","down","error","systematic","me","document","bug","use","fix","tracking","test","read","work","reproduce","why","verify","help","systematic debugging","debug","2026-11-10 10:12:30",false,"---\nname: debugging\ndescription: |\n  Systematic debugging — reproduce, read the error, isolate, hypothesize, fix, verify, document. Use when tracking down a bug, unexpected behavior, or test failure — including phrases like \"this doesn't work\", \"why is this failing\", or \"help me debug\".\nmetadata:\n  processkit:\n    apiVersion: processkit.projectious.work\u002Fv2\n    id: SKILL-debugging\n    version: \"1.1.0\"\n    created: 2026-04-06T00:00:00Z\n    category: engineering\n    layer: 3\n---\n\n# Debugging\n\n## Intro\n\nDebugging is a fixed loop: reproduce the failure, read the error,\nisolate the cause, fix it, verify the fix, and add a regression\ntest. Skipping steps causes the bug to come back.\n\n## Overview\n\n### The debugging loop\n\n1. **Reproduce** — get the exact steps, inputs, and error\n   messages. Confirm you can trigger the issue on demand.\n2. **Read the error** — parse the full stack trace or error\n   message. The root cause is usually in the first or last frame,\n   not the middle.\n3. **Isolate** — narrow the scope:\n   - Binary search through code (comment out halves)\n   - Check recent changes (`git log`, `git diff`)\n   - Add targeted logging or print statements\n   - Test with the minimal failing input\n4. **Hypothesize and test** — form a theory, then test it. Change\n   one thing at a time.\n5. **Fix** — apply the minimal change that resolves the issue.\n6. **Verify** — confirm the original error is gone, run the full\n   test suite for regressions, and exercise edge cases near the\n   fix.\n7. **Document** — add a test that would have caught this bug, and\n   a comment if the fix is non-obvious.\n\n### Common diagnosis paths\n\n- **Tests pass locally, fail in CI** — environment differences:\n  OS, dependency versions, file paths, timezone, locale, or race\n  conditions in parallel tests. Compare CI logs with local output\n  to find the divergence point.\n- **Function returns wrong results for some inputs** — get the\n  specific failing inputs, add assertions at intermediate steps,\n  find where actual diverges from expected, trace back.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Changing multiple things at once.** When more than one variable changes between a broken and fixed state, you cannot know which change fixed the issue — and you may have introduced new bugs alongside the fix. Change one thing at a time.\n- **Skipping the reproduction step.** Starting to diagnose before having a reliable reproduction means you're hunting in the dark. Confirm you can trigger the failure on demand before doing anything else.\n- **Stopping at the first fix that makes the symptom disappear.** The first change that silences the error may be masking the root cause. Verify that the underlying state is correct, not just the visible output.\n- **No regression test after the fix.** A bug that was fixed without a test will return. Add a test that would have caught it before closing the issue.\n- **Adding print statements everywhere without a plan.** Each print should test a specific hypothesis (\"is the value non-nil here?\"). Random prints produce noise that slows diagnosis.\n- **Confusing correlation with causation in logs.** The last log entry before a crash is rarely the cause — it is just the last thing that ran. Trace the full causal chain; do not assume temporal proximity implies causation.\n- **Treating a CI\u002Flocal mismatch as mysterious.** Environment differences (OS, timezone, locale, implicit tool version, parallel test ordering) are the most common cause. Diff the CI environment against local systematically rather than guessing.\n\n## Full reference\n\n### Anti-patterns\n\n- **Shotgun debugging** — changing many things at once. You won't\n  know which fix worked, and you'll introduce new bugs.\n- **Print-statement spelunking with no plan** — adding prints is\n  fine, but each one should test a specific hypothesis.\n- **Skipping reproduction** — \"it sometimes fails\" is not a bug\n  report. Find the trigger first.\n- **Stopping at the first fix** — the first thing that makes the\n  symptom go away is often masking the real cause. Verify the\n  underlying state is correct, not just the visible output.\n- **No regression test** — if you don't add a test, the bug will\n  return.\n\n### Tools and techniques\n\n- **`git bisect`** for regressions: a known-good and known-bad\n  commit narrows the offending change in log(N) steps.\n- **Debuggers over prints** when state is rich (object graphs,\n  step-through control flow).\n- **Differential debugging** — diff the working case against the\n  failing case, not the other way around.\n- **Rubber-duck explanation** — articulating the problem out loud\n  often surfaces the bug before you write any code.\n\n### When to stop and ask\n\nIf you've spent more than ~30 minutes without forming a viable\nhypothesis, stop and either gather more information (logs,\nreproduction steps, recent commits) or escalate. Continued flailing\nis not progress.\n","# Debugging\n\n## Intro\n\nDebugging is a fixed loop: reproduce the failure, read the error,\nisolate the cause, fix it, verify the fix, and add a regression\ntest. Skipping steps causes the bug to come back.\n\n## Overview\n\n### The debugging loop\n\n1. **Reproduce** — get the exact steps, inputs, and error\n   messages. Confirm you can trigger the issue on demand.\n2. **Read the error** — parse the full stack trace or error\n   message. The root cause is usually in the first or last frame,\n   not the middle.\n3. **Isolate** — narrow the scope:\n   - Binary search through code (comment out halves)\n   - Check recent changes (`git log`, `git diff`)\n   - Add targeted logging or print statements\n   - Test with the minimal failing input\n4. **Hypothesize and test** — form a theory, then test it. Change\n   one thing at a time.\n5. **Fix** — apply the minimal change that resolves the issue.\n6. **Verify** — confirm the original error is gone, run the full\n   test suite for regressions, and exercise edge cases near the\n   fix.\n7. **Document** — add a test that would have caught this bug, and\n   a comment if the fix is non-obvious.\n\n### Common diagnosis paths\n\n- **Tests pass locally, fail in CI** — environment differences:\n  OS, dependency versions, file paths, timezone, locale, or race\n  conditions in parallel tests. Compare CI logs with local output\n  to find the divergence point.\n- **Function returns wrong results for some inputs** — get the\n  specific failing inputs, add assertions at intermediate steps,\n  find where actual diverges from expected, trace back.\n\n## Gotchas\n\nAgent-specific failure modes — provider-neutral pause-and-self-check items:\n\n- **Changing multiple things at once.** When more than one variable changes between a broken and fixed state, you cannot know which change fixed the issue — and you may have introduced new bugs alongside the fix. Change one thing at a time.\n- **Skipping the reproduction step.** Starting to diagnose before having a reliable reproduction means you're hunting in the dark. Confirm you can trigger the failure on demand before doing anything else.\n- **Stopping at the first fix that makes the symptom disappear.** The first change that silences the error may be masking the root cause. Verify that the underlying state is correct, not just the visible output.\n- **No regression test after the fix.** A bug that was fixed without a test will return. Add a test that would have caught it before closing the issue.\n- **Adding print statements everywhere without a plan.** Each print should test a specific hypothesis (\"is the value non-nil here?\"). Random prints produce noise that slows diagnosis.\n- **Confusing correlation with causation in logs.** The last log entry before a crash is rarely the cause — it is just the last thing that ran. Trace the full causal chain; do not assume temporal proximity implies causation.\n- **Treating a CI\u002Flocal mismatch as mysterious.** Environment differences (OS, timezone, locale, implicit tool version, parallel test ordering) are the most common cause. Diff the CI environment against local systematically rather than guessing.\n\n## Full reference\n\n### Anti-patterns\n\n- **Shotgun debugging** — changing many things at once. You won't\n  know which fix worked, and you'll introduce new bugs.\n- **Print-statement spelunking with no plan** — adding prints is\n  fine, but each one should test a specific hypothesis.\n- **Skipping reproduction** — \"it sometimes fails\" is not a bug\n  report. Find the trigger first.\n- **Stopping at the first fix** — the first thing that makes the\n  symptom go away is often masking the real cause. Verify the\n  underlying state is correct, not just the visible output.\n- **No regression test** — if you don't add a test, the bug will\n  return.\n\n### Tools and techniques\n\n- **`git bisect`** for regressions: a known-good and known-bad\n  commit narrows the offending change in log(N) steps.\n- **Debuggers over prints** when state is rich (object graphs,\n  step-through control flow).\n- **Differential debugging** — diff the working case against the\n  failing case, not the other way around.\n- **Rubber-duck explanation** — articulating the problem out loud\n  often surfaces the bug before you write any code.\n\n### When to stop and ask\n\nIf you've spent more than ~30 minutes without forming a viable\nhypothesis, stop and either gather more information (logs,\nreproduction steps, recent commits) or escalate. Continued flailing\nis not progress.\n",{"name":5,"metadata":43,"description":51},{"processkit":44},{"id":45,"layer":46,"created":47,"version":48,"category":49,"apiVersion":50},"SKILL-debugging",3,1775433600000,"1.1.0","engineering","processkit.projectious.work\u002Fv2","Systematic debugging — reproduce, read the error, isolate, hypothesize, fix, verify, document. Use when tracking down a bug, unexpected behavior, or test failure — including phrases like \"this doesn't work\", \"why is this failing\", or \"help me debug\".\n","main","7231f100e266e30d2c1699ecde5c00602471dc3c","src\u002Fcontext\u002Fskills\u002Fengineering\u002Fdebugging","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit","https:\u002F\u002Fprojectious-work.github.io\u002Fprocesskit\u002F","https:\u002F\u002Fgithub.com\u002Fprojectious-work","2026-11-11 23:24:18",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add projectious-work\u002Fprocesskit --skill debugging",{"tool":65,"command":66},"bunx","bunx skills add projectious-work\u002Fprocesskit --skill debugging",{"tool":68,"command":69},"pnpm","pnpm dlx skills add projectious-work\u002Fprocesskit --skill debugging","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fengineering\u002Fdebugging\u002FSKILL.md",true,[73],{"path":74,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":14,"sourceHtmlUrl":77},"scripts\u002F.gitkeep","SCRIPT","脚本","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fengineering\u002Fdebugging\u002Fscripts\u002F.gitkeep",[79,86,93,99,126,146],{"skillNo":80,"name":81,"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":82,"categories":83,"keywords":84,"publishedAt":85},"01M7665G2ED3624R3X93304265","standard-ui-design","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-ui-design\u002FSKILL.md",[],[],"2026-11-12 13:55:43",{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":85},"01M76651QTC6W0HJ4AFW2MMR3V","standard-agent-use","Company standard for working with AI agents. Use when deciding what to delegate to agents, capturing open work, keeping agent memory or knowledge notes, improving an agent workflow, or introducing agent work in a team or for a client.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-use\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"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":96,"categories":97,"keywords":98,"publishedAt":85},"01M76650NN4T8P241021NBR64S","standard-agent-building","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fstandard-agent-building\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":110,"publishedAt":38},"01M6X5MTC89B4ZGXW3JVSQCPF8","skill-management","Manage Skill package manifests as v1 ontology entities.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fskill-management\u002FSKILL.md",[105],{"categoryNo":106,"categoryName":107,"taxonomyType":108,"taxonomyTypeDesc":109},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[111,112,113,114,115,116,31,117,118,119,120,121,101,122,123,124,125],"management","skill management","ontology","v1","create","as","transition","get","manage","update","alpha","list","package","skill package","manifests",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":133,"publishedAt":38},"01M6X5MCV74NHW5MEDNHZRF2C8","schema-management","Manage generated processkit schema contracts through deterministic MCP rebuild and inspection workflows.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fschema-management\u002FSKILL.md",[132],{"categoryNo":106,"categoryName":107,"taxonomyType":108,"taxonomyTypeDesc":109},[134,135,111,136,137,138,139,140,141,142,143,144,31,118,119,121,145],"schema","schema management","contract","rebuild","contracts","workflows","wed","generated","MCP","validation","deterministic","mode",{"skillNo":147,"name":148,"nameZh":6,"description":149,"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":38},"01M6X5M5G6YJ96AMZGSCZJ1D13","proposition-management","Record claims, risks, beliefs, world facts, WSJF estimates, and assumptions through the Proposition interface.","https:\u002F\u002Fgithub.com\u002Fprojectious-work\u002Fprocesskit\u002Fblob\u002Fmain\u002Fsrc\u002Fcontext\u002Fskills\u002Fprocesskit\u002Fproposition-management\u002FSKILL.md",[],[153,111,154,155,115,156,157,158,31,118,159,160,120,161,121,162,163,164,165,166],"proposition management","proposition","interface","claims","record","world","creation","event","WSJF","index","updates","query","facts","log"]