[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CP1EWMXF37SAQ58ZJQQ85H":3,"skill-same:01M7CP1EWMXF37SAQ58ZJQQ85H":56},{"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":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M7CP1EWMXF37SAQ58ZJQQ85H","perf-issue",null,"Measured performance work. Use when the user asks to make something faster or use less memory or CPU, or asks why it is slow.","syanga\u002Fastack","syanga","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F42523849?v=4","01M7CNYD2SW40HXC8R8CTVJPGE","Alan Yang","01M7CNYD3D0B0EYPA4AGXC1VC8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fperf-issue\u002FLICENSE-pstack","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fperf-issue\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: perf-issue\ndescription: Measured performance work. Use when the user asks to make something faster or use less memory or CPU, or asks why it is slow.\n---\n\n# Perf issue\n\n**You own the measurements. Plan, review, verify the numbers.** Tie every fix to a measurement.\n\nWhen the user asks for sustained improvement against a metric, such as iterating toward a target, use [hillclimb](..\u002Fhillclimb\u002FSKILL.md). Pass the workload, target, baseline evidence, traced model, measurement command, correctness checks, delivery scope, and existing authorization. Hillclimb then owns the repeated experiments, stopping criteria, and delivery.\n\nFor a diagnosis-only request, stop before implementation with the measurements, supported cause, and proposed next experiment. A read-only request also excludes instrumentation and code edits.\n\n0. Read [scale the workflow](..\u002Fprinciples\u002Fscale-the-workflow.md) before anything else in this skill.\n1. Reproduce the reported problem in an authorized environment and capture a baseline through the affected UI, CLI, or API. Name the metric, unit, target, workload, revision, configuration, and repeatable command. Without a target from the user, use the metric value at which the complaint no longer reproduces. Record warmup, trial count, and variation so later comparisons can distinguish a change from noise. Vet the baseline, and each later number, with the [benchmark checklist](BENCHMARK-CHECKLIST.md). When the fix under test is a setting, such as an index or batching, keep that setting at production's current value on the baseline, and tune every other setting. For live capture and mechanism checks, follow [runtime forensics](..\u002Fdiagnosing-bugs\u002FRUNTIME-FORENSICS.md). For an existing capture, follow [trace forensics](..\u002Fdiagnosing-bugs\u002FTRACE-FORENSICS.md). For a diagnosis-only request, the capture can replace the live baseline. For a regression, also measure a known-good revision, and bisect to the change that slowed it. If the baseline does not reproduce the complaint, report the gap before optimizing.\n2. Use [how](..\u002Fhow\u002FSKILL.md) to ground hypotheses, unless you already hold a traced model of the hot path. Measure the workload before claiming a performance limit.\n   Most fixes come from eight strategy families. Try them in order, cheapest first. A family earns an attempt only when the trace shows the signal it names. When an earlier family meets the target, stop.\n   1. **Elimination.** Before optimizing the hot path, ask whether it needs to exist: a computation nobody consumes, a feature gate that's always off for this user, a sync that redundantly mirrors state, a legacy path kept \"just in case\". The trace shows what's slow, never that it's deletable, so this family needs the traced model from this step, not the profiler.\n   2. **Caching.** The same computation or fetch repeats on identical inputs. Store and reuse the result. Name what invalidates it before claiming the win.\n   3. **Batching.** Many small operations each pay a fixed overhead (RPC, query, syscall, draw call). Coalesce them to pay the overhead once per batch.\n   4. **Lazy evaluation.** Cost lands on results that are never used or not needed yet (eager init on the boot path, rendering offscreen items). Defer the work until first use.\n   5. **Scheduling.** The work must happen, but not during the interactive moment. Move it to where nobody is waiting: idle callbacks, a background warmup after boot, precompute before the user arrives, cleanup after the frame commits. The win is perceived latency, so measure the interactive path, not total work done.\n   6. **Divide and conquer.** The dominant cost scales with input size. Split the work so each piece touches less (chunk, shard, prune the search space) or so independent pieces run in parallel.\n   7. **Redundancy.** The wait hangs on one slow instance or attempt. Duplicate the work (replicas, hedged requests, speculative execution) and take the fastest result. The trace has to show the wait dominates and the system has headroom.\n   8. **Indirection.** The hot path does expensive work a cheaper intermediate could absorb: an index instead of a scan, a queue that shifts work off the interactive thread, a handle that lets a cheaper implementation swap in. Add the hop only when it removes more from the critical path than it adds.\n3. Plan the fix from the measurements. If it crosses a function boundary, read [architect](..\u002Farchitect\u002FSKILL.md) first. When you delegate the fix, give the worker the hypothesis, scope, fixed workload, measurement command, and correctness checks. Review the diff and capture post-fix measurements.\n   Apply [sequence-verifiable-units](..\u002Fprinciples\u002Fsequence-verifiable-units.md), verifying each attempt before trying the next. Remove rejected experimental changes while preserving unrelated work.\n4. Parse and compare the artifacts using the same workload, environment, metric, and measurement method. Change the code revision under test, and record both revisions. Report the absolute and percentage delta with the observed variation. Run correctness checks, using [tdd](..\u002Ftdd\u002FSKILL.md) when adding a behavioral regression test. An inconclusive measurement or a measurement of a different interaction is not a pass. Remove temporary instrumentation or record why it remains and who owns cleanup.\n5. End with [review-pr](..\u002Freview-pr\u002FSKILL.md) in fix-and-review mode against the revision the fix started from. Run it for every fix, however small. Its reviewers are independent of you, so reviewing the diff yourself does not replace it. When the task's delivery scope includes a PR, use [open-pr](..\u002Fopen-pr\u002FSKILL.md) instead, which ends with review-pr. When a PR exists, cite the commands, results, and artifact paths under its `## Verification`. Merging is a separate action.\n\nFor a unit in a multi-phase plan, follow [the shared execution-state convention](..\u002Forchestrate\u002FSTATE.md). Reuse its store. For a session transfer, read [handoff](..\u002Fhandoff\u002FSKILL.md) and include the measurement command, workload, results, and artifact paths.\n\n**Reply:** verdict, metric, baseline number, post-fix number, delta, variation and run count, limiter, whether the target was met, correctness checks, and artifact paths.\n","# Perf issue\n\n**You own the measurements. Plan, review, verify the numbers.** Tie every fix to a measurement.\n\nWhen the user asks for sustained improvement against a metric, such as iterating toward a target, use [hillclimb](..\u002Fhillclimb\u002FSKILL.md). Pass the workload, target, baseline evidence, traced model, measurement command, correctness checks, delivery scope, and existing authorization. Hillclimb then owns the repeated experiments, stopping criteria, and delivery.\n\nFor a diagnosis-only request, stop before implementation with the measurements, supported cause, and proposed next experiment. A read-only request also excludes instrumentation and code edits.\n\n0. Read [scale the workflow](..\u002Fprinciples\u002Fscale-the-workflow.md) before anything else in this skill.\n1. Reproduce the reported problem in an authorized environment and capture a baseline through the affected UI, CLI, or API. Name the metric, unit, target, workload, revision, configuration, and repeatable command. Without a target from the user, use the metric value at which the complaint no longer reproduces. Record warmup, trial count, and variation so later comparisons can distinguish a change from noise. Vet the baseline, and each later number, with the [benchmark checklist](BENCHMARK-CHECKLIST.md). When the fix under test is a setting, such as an index or batching, keep that setting at production's current value on the baseline, and tune every other setting. For live capture and mechanism checks, follow [runtime forensics](..\u002Fdiagnosing-bugs\u002FRUNTIME-FORENSICS.md). For an existing capture, follow [trace forensics](..\u002Fdiagnosing-bugs\u002FTRACE-FORENSICS.md). For a diagnosis-only request, the capture can replace the live baseline. For a regression, also measure a known-good revision, and bisect to the change that slowed it. If the baseline does not reproduce the complaint, report the gap before optimizing.\n2. Use [how](..\u002Fhow\u002FSKILL.md) to ground hypotheses, unless you already hold a traced model of the hot path. Measure the workload before claiming a performance limit.\n   Most fixes come from eight strategy families. Try them in order, cheapest first. A family earns an attempt only when the trace shows the signal it names. When an earlier family meets the target, stop.\n   1. **Elimination.** Before optimizing the hot path, ask whether it needs to exist: a computation nobody consumes, a feature gate that's always off for this user, a sync that redundantly mirrors state, a legacy path kept \"just in case\". The trace shows what's slow, never that it's deletable, so this family needs the traced model from this step, not the profiler.\n   2. **Caching.** The same computation or fetch repeats on identical inputs. Store and reuse the result. Name what invalidates it before claiming the win.\n   3. **Batching.** Many small operations each pay a fixed overhead (RPC, query, syscall, draw call). Coalesce them to pay the overhead once per batch.\n   4. **Lazy evaluation.** Cost lands on results that are never used or not needed yet (eager init on the boot path, rendering offscreen items). Defer the work until first use.\n   5. **Scheduling.** The work must happen, but not during the interactive moment. Move it to where nobody is waiting: idle callbacks, a background warmup after boot, precompute before the user arrives, cleanup after the frame commits. The win is perceived latency, so measure the interactive path, not total work done.\n   6. **Divide and conquer.** The dominant cost scales with input size. Split the work so each piece touches less (chunk, shard, prune the search space) or so independent pieces run in parallel.\n   7. **Redundancy.** The wait hangs on one slow instance or attempt. Duplicate the work (replicas, hedged requests, speculative execution) and take the fastest result. The trace has to show the wait dominates and the system has headroom.\n   8. **Indirection.** The hot path does expensive work a cheaper intermediate could absorb: an index instead of a scan, a queue that shifts work off the interactive thread, a handle that lets a cheaper implementation swap in. Add the hop only when it removes more from the critical path than it adds.\n3. Plan the fix from the measurements. If it crosses a function boundary, read [architect](..\u002Farchitect\u002FSKILL.md) first. When you delegate the fix, give the worker the hypothesis, scope, fixed workload, measurement command, and correctness checks. Review the diff and capture post-fix measurements.\n   Apply [sequence-verifiable-units](..\u002Fprinciples\u002Fsequence-verifiable-units.md), verifying each attempt before trying the next. Remove rejected experimental changes while preserving unrelated work.\n4. Parse and compare the artifacts using the same workload, environment, metric, and measurement method. Change the code revision under test, and record both revisions. Report the absolute and percentage delta with the observed variation. Run correctness checks, using [tdd](..\u002Ftdd\u002FSKILL.md) when adding a behavioral regression test. An inconclusive measurement or a measurement of a different interaction is not a pass. Remove temporary instrumentation or record why it remains and who owns cleanup.\n5. End with [review-pr](..\u002Freview-pr\u002FSKILL.md) in fix-and-review mode against the revision the fix started from. Run it for every fix, however small. Its reviewers are independent of you, so reviewing the diff yourself does not replace it. When the task's delivery scope includes a PR, use [open-pr](..\u002Fopen-pr\u002FSKILL.md) instead, which ends with review-pr. When a PR exists, cite the commands, results, and artifact paths under its `## Verification`. Merging is a separate action.\n\nFor a unit in a multi-phase plan, follow [the shared execution-state convention](..\u002Forchestrate\u002FSTATE.md). Reuse its store. For a session transfer, read [handoff](..\u002Fhandoff\u002FSKILL.md) and include the measurement command, workload, results, and artifact paths.\n\n**Reply:** verdict, metric, baseline number, post-fix number, delta, variation and run count, limiter, whether the target was met, correctness checks, and artifact paths.\n",{"name":5,"description":7},"main","ec1ea5dc71f382a2a09360a736d31b7294e5ecc2","skills\u002Fperf-issue","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack","https:\u002F\u002Fgithub.com\u002Fsyanga","2026-11-14 17:01:29",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add syanga\u002Fastack --skill perf-issue",{"tool":37,"command":38},"bunx","bunx skills add syanga\u002Fastack --skill perf-issue",{"tool":40,"command":41},"pnpm","pnpm dlx skills add syanga\u002Fastack --skill perf-issue","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fperf-issue\u002FSKILL.md",true,[45,51],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"BENCHMARK-CHECKLIST.md","REFERENCE","参考文档",4439,"https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fperf-issue\u002FBENCHMARK-CHECKLIST.md",{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":16},"LICENSE-pstack","LICENSE","许可证",1067,[57,64,72,80,88,96],{"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":20},"01M7CP39S6D7662SFCC9K286T7","replace-with-skill-name","Replace with when this skill should be reached.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Ftemplates\u002Fskill\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":68,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":20},"01M7CP37YN62Z7VR6P2J7918ZA","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":76,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":20},"01M7CP353RP2S1M1ZMHFZ0F610","worktree-cleanup","Worktree cleanup. Use for cleanup, resource audits, close this lane, or close out.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fworktree-cleanup\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fworktree-cleanup\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":84,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":20},"01M7CP32DSVPRTBAFHF1P6NZ1X","wizard","Generate an interactive Bash wizard that walks a human through steps only they can perform. Use when provisioning, credential setup, or a one-off migration needs such steps.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwizard\u002FLICENSE-mattpocock","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwizard\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":92,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":20},"01M7CP2YE3WABQE9MNY8RQD6MX","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Returns a cited read on decisions and tradeoffs.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwhy\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwhy\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":100,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":20},"01M7CP2VP8JA1HYDE8TJX2Y494","wayfinder","Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwayfinder\u002FLICENSE-mattpocock","https:\u002F\u002Fgithub.com\u002Fsyanga\u002Fastack\u002Fblob\u002Fmain\u002Fskills\u002Fwayfinder\u002FSKILL.md",[],[]]