[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CMM71NSJ4QNTGVADBQCSDD":3,"skill-same:01M7CMM71NSJ4QNTGVADBQCSDD":132},{"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":30,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M7CMM71NSJ4QNTGVADBQCSDD","unity-profiling",null,"Measure frame time, hitches and GC with ProfilerRecorder counters; before\u002Fafter numbers.","tipoLi5890\u002Funity-dev-skills","tipoLi5890","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F177803612?v=4","01M7CMD7DHGZ09CB53CRFHAA34","Li, ching yu","01M7CMD7E70K5RAHWWZVYX9YMG",2,"MIT","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: unity-profiling\ndescription: >-\n  Measure frame time, hitches and GC with ProfilerRecorder counters; before\u002Fafter numbers.\n---\n\n# unity-profiling — measure the work, then compare the table\n\n\"It feels smoother now\" survives exactly until someone else plays it. A frame time on its own is\nbarely better: one number, from one moment, on one machine, with nothing to hold it against.\n\n## Load what the situation needs\n\n| The situation | Read |\n|---|---|\n| Starting from nothing; \"how many draw calls is this\"; \"which counters exist on my version\"; a counter reads zero | `reference\u002Fcounters.md` |\n| \"The frame rate drops when…\" — taking the measurement: warm-up, window size, run modes, the log line and the JSONL schema | `reference\u002Fframe-windows.md` |\n| \"It hitches when the third pickup spawns\" — hitches around spawning, streaming or pooling; a system that builds the same thing twice | `reference\u002Fwork-counters.md` |\n| \"did my change actually help\"; building the before\u002Fafter table; a comparison that refuses | `reference\u002Fcompare.md` |\n| A window in hand and no cause yet: first-appearance hitch, GC, synchronous loads, distant geometry | `reference\u002Fwhere-the-time-goes.md` |\n| No counter names the cost and you need the marker inside the frame | `reference\u002Fraw-capture.md` |\n| A frame-time budget that a test fails on, every run, against this machine's recorded baseline | `reference\u002Fperf-tests.md` · `templates\u002FPerfRegressionTest.cs` |\n| \"is it the GC\" and whose: allocation calls per Update and LateUpdate phase; `GetAllocatedBytesForCurrentThread` reads 0 | `reference\u002Fgc-per-phase.md` · `resources\u002FGcAllocProbe.cs` |\n| Memory grows by a step each time a screen or match is rebuilt; runtime materials and meshes | `reference\u002Fruntime-asset-leaks.md` · `resources\u002FOwnedAssets.cs` · `templates\u002FAssetLeakSoakTest.cs` |\n| A frame time read on a shared machine; post-processing cost; CPU or GPU in the Editor | `reference\u002Fframe-hygiene.md` |\n\n## Files and instruments\n\nFiles to copy into `Assets\u002F` — namespace `UnityDev.Perf`, Runtime, no package required:\n`resources\u002FCounterCatalog.cs`, `resources\u002FFrameSampler.cs`, `resources\u002FWorkCounters.cs`,\n`resources\u002FProfilerRawCapture.cs`, plus `resources\u002FGcAllocProbe.cs` with `GcAllocProbeOpener.cs`\nfor allocation by phase and `resources\u002FOwnedAssets.cs` for runtime-asset ownership (that one is\ngameplay code). The rest are instrumentation, not gameplay; keep them behind a\ndevelopment-build check before shipping.\n\nWhat settles a performance question is a **window** — a fixed number of frames of a named\nsituation, summarised into one row — taken twice, with one thing changed in between.\n\n> **A driven run is not a hardware benchmark.**\n> `batchmode-nographics`, `editor-windowed` and a player build are three different instruments.\n> Each is internally consistent and none of their numbers mean anything against another's. Every\n> window records which instrument produced it, and the comparison script refuses to subtract\n> across them. And a human playing the actual build still decides shader compilation, GPU upload\n> and whether the thing feels good — measurement narrows the question, it does not answer it.\n\n---\n\n## The loop\n\n**1 · List the counters this build exposes.** Never recall them. Names differ by render pipeline,\ngraphics backend, platform and version, and a wrong name produces a recorder that reads `0` —\nindistinguishable from \"this build draws nothing\". `resources\u002FCounterCatalog.cs`, then\n`reference\u002Fcounters.md`.\n\n**2 · Sample a window.** 90 frames after 30 warm-up frames, sampled in `LateUpdate` at the end of\nthe execution order with a `Stopwatch` as the clock. Out comes one greppable `[PERF]` line and one\nJSONL record. `resources\u002FFrameSampler.cs`, then `reference\u002Fframe-windows.md`.\n\n**3 · Change exactly one thing.** Two changes in one window cannot be attributed, and the wrong\none gets kept.\n\n**4 · Re-run the same situation and compare.** Same seed, same instrument, three windows a side.\n`scripts\u002Fcompare_windows.py`, then `reference\u002Fcompare.md`.\n\n### One turn of the loop\n\nThe complaint is *\"it hitches when the third pickup spawns in the hazard lane\"*. Booted into that\nsituation with a fixed seed, one window says:\n\n```\n[PERF] window=hazard-lane mode=editor-windowed frames=90 ft_med=8.31 ft_p95=12.90 ft_max=41.2\n       hitches=2 draw_med=412 tris_med=51200 gc_sum=184320 disc[pickup-pool]=214 missing=1 zero=0\n```\n\nRead it in that order. The median is fine, so the game is not uniformly heavy; `ft_p95` at 12.90\nagainst a median of 8.31 is unevenness. Draw calls and triangles are the same as the run before,\nso it is not a rendering cost at all. The last number is the finding: the pickup pool **threw away\n214 units of work** during those ninety frames while completing seventeen. The lane selector is\nflipping across a 1.3-unit lane boundary and cancelling the set it was building each time.\n\nA 0.2-unit dead band on that boundary, then the same window again: discarded 214 → 7, `ft_p95`\n12.90 → 9.10 ms, hitches 2 → 0, triangles unchanged. One parameter, two numbers that moved\ntogether, and a table that says why.\n\n---\n\n## Where the time goes\n\nA router, read against a window you already have. Depth for the first four rows is in\n`reference\u002Fwhere-the-time-goes.md`; the rest belong to the skill that owns the subsystem.\n\n| The window says | Go to |\n|---|---|\n| One large `ft_max` the first time something appears, never again | first-appearance section, then `unity-render-urp` |\n| `Memory\u002FGC Allocated In Frame` has a non-zero median; spikes on a period | GC section, then `unity-game-ui` for per-frame string building |\n| A spike lined up with a jump in a completed-work counter | synchronous-loads section, then `unity-audio` when it is a clip |\n| Triangles high while looking at distance, draw calls ordinary | distant-geometry section, then `unity-scene-optimisation` |\n| Discarded work climbing while scheduled climbs | `reference\u002Fwork-counters.md` |\n| Draw calls and SetPass calls high, triangles ordinary | `unity-scene-optimisation` · `unity-sprite-atlas` for 2D · `unity-game-ui` for canvases |\n| GPU frame time much larger than the CPU times | `unity-scene-optimisation` · `unity-render-urp` for post-processing |\n| Shadow caster count high | `unity-lighting` |\n| Pixel art shimmering as well as slow | `unity-2d-pixel-perfect` |\n| Slow only in the browser | `unity-web-release` |\n| Slow only on the device, fine in the editor | `unity-device-testing`, after a development-player window taken on it |\n| Frame time is fine and it still feels wrong | `unity-play-harness` |\n\n---\n\n## Discipline\n\n- **List, never recall.** Counter names are defined by the build, not by your memory of them. One\n  catalogue call is cheaper than one wrong name, and a wrong name reports zero rather than failing.\n- **A counter that reads zero on every frame of a window is an absence, not a measurement of\n  nothing.** The window separates the two: `missing[]` is \"not exposed on this build\",\n  `zeroAll[]` is \"exposed and flat\". Report both; never quietly average a zero into a result.\n- **A quiet machine, or do not measure.** Another Editor or a Blender job turned 1.6–2.4 ms into\n  4–10 ms in one project; check the process list first (`reference\u002Fframe-hygiene.md`).\n- **Same situation, same instrument, or do not compare.** Different run mode, graphics device,\n  screen size, vsync or warm-up makes two numbers that subtract cleanly and mean nothing.\n- **Median and worst, never just worst** — one spike and a constant vibration are different faults\n  — with p95 between them to say where the bad frames start.\n- **Warm up 30 frames by default.** The first frames after entering a situation measure starting,\n  not playing. Set warm-up to 0 deliberately when a first-appearance hitch *is* the subject, and\n  put a `-cold` suffix on that window's label so nobody compares it against a warmed one.\n- **Three windows a side before believing a 10% change.** Below that, one window each way is a\n  coin toss, and the comparison script will tell you when the spread makes the change unreadable.\n- **Delete the JSONL before every run.** It is appended to. A stale window compared against a\n  fresh one is the quietest wrong answer this skill can produce.\n- **Print the numbers the judgement rests on.** \"The hitch is gone\" ages badly; \"discarded 214 → 7,\n  p95 12.90 → 9.10 ms\" keeps saying the same thing after the next change.\n- **Check whether the candidate is faster because it does less.** Fewer things built is not the\n  same as things built more cheaply. Read the scheduled and completed counters before celebrating.\n- **Every report names its run mode and what it did not cover.** A report that omits the\n  instrument invites the reader to assume it was the shipped game.\n\n## Scope — what this skill does NOT do\n\n| Not here | Go to |\n|---|---|\n| Booting into a moment, a readable state line, a feel complaint with nothing to measure | `unity-play-harness` |\n| Driving Unity, exit codes, screenshots, proving a harness told the truth | `unity-debug` |\n| Installing editors, licences, CI invocation | `unity-cli` |\n| Atlasing sprites, decimating meshes | `unity-sprite-atlas` · `unity-3d-models` |\n| Configuring post-processing and renderer features | `unity-render-urp` |\n| Browser download size, headers and memory ceilings | `unity-web-release` |\n| Building, signing and reading a device artifact back | `unity-android-release` |\n| Stutters on the device, runs fine in the Editor — a window taken on the phone | `unity-device-testing` |\n| Audio memory, load types and mixer cost | `unity-audio` |\n| Deciding what the budget should be | `unity-game-brief` |\n| Where a character's frame windows fall: the budget ladder, the import check | `game-character-pipeline` |\n","# unity-profiling — measure the work, then compare the table\n\n\"It feels smoother now\" survives exactly until someone else plays it. A frame time on its own is\nbarely better: one number, from one moment, on one machine, with nothing to hold it against.\n\n## Load what the situation needs\n\n| The situation | Read |\n|---|---|\n| Starting from nothing; \"how many draw calls is this\"; \"which counters exist on my version\"; a counter reads zero | `reference\u002Fcounters.md` |\n| \"The frame rate drops when…\" — taking the measurement: warm-up, window size, run modes, the log line and the JSONL schema | `reference\u002Fframe-windows.md` |\n| \"It hitches when the third pickup spawns\" — hitches around spawning, streaming or pooling; a system that builds the same thing twice | `reference\u002Fwork-counters.md` |\n| \"did my change actually help\"; building the before\u002Fafter table; a comparison that refuses | `reference\u002Fcompare.md` |\n| A window in hand and no cause yet: first-appearance hitch, GC, synchronous loads, distant geometry | `reference\u002Fwhere-the-time-goes.md` |\n| No counter names the cost and you need the marker inside the frame | `reference\u002Fraw-capture.md` |\n| A frame-time budget that a test fails on, every run, against this machine's recorded baseline | `reference\u002Fperf-tests.md` · `templates\u002FPerfRegressionTest.cs` |\n| \"is it the GC\" and whose: allocation calls per Update and LateUpdate phase; `GetAllocatedBytesForCurrentThread` reads 0 | `reference\u002Fgc-per-phase.md` · `resources\u002FGcAllocProbe.cs` |\n| Memory grows by a step each time a screen or match is rebuilt; runtime materials and meshes | `reference\u002Fruntime-asset-leaks.md` · `resources\u002FOwnedAssets.cs` · `templates\u002FAssetLeakSoakTest.cs` |\n| A frame time read on a shared machine; post-processing cost; CPU or GPU in the Editor | `reference\u002Fframe-hygiene.md` |\n\n## Files and instruments\n\nFiles to copy into `Assets\u002F` — namespace `UnityDev.Perf`, Runtime, no package required:\n`resources\u002FCounterCatalog.cs`, `resources\u002FFrameSampler.cs`, `resources\u002FWorkCounters.cs`,\n`resources\u002FProfilerRawCapture.cs`, plus `resources\u002FGcAllocProbe.cs` with `GcAllocProbeOpener.cs`\nfor allocation by phase and `resources\u002FOwnedAssets.cs` for runtime-asset ownership (that one is\ngameplay code). The rest are instrumentation, not gameplay; keep them behind a\ndevelopment-build check before shipping.\n\nWhat settles a performance question is a **window** — a fixed number of frames of a named\nsituation, summarised into one row — taken twice, with one thing changed in between.\n\n> **A driven run is not a hardware benchmark.**\n> `batchmode-nographics`, `editor-windowed` and a player build are three different instruments.\n> Each is internally consistent and none of their numbers mean anything against another's. Every\n> window records which instrument produced it, and the comparison script refuses to subtract\n> across them. And a human playing the actual build still decides shader compilation, GPU upload\n> and whether the thing feels good — measurement narrows the question, it does not answer it.\n\n---\n\n## The loop\n\n**1 · List the counters this build exposes.** Never recall them. Names differ by render pipeline,\ngraphics backend, platform and version, and a wrong name produces a recorder that reads `0` —\nindistinguishable from \"this build draws nothing\". `resources\u002FCounterCatalog.cs`, then\n`reference\u002Fcounters.md`.\n\n**2 · Sample a window.** 90 frames after 30 warm-up frames, sampled in `LateUpdate` at the end of\nthe execution order with a `Stopwatch` as the clock. Out comes one greppable `[PERF]` line and one\nJSONL record. `resources\u002FFrameSampler.cs`, then `reference\u002Fframe-windows.md`.\n\n**3 · Change exactly one thing.** Two changes in one window cannot be attributed, and the wrong\none gets kept.\n\n**4 · Re-run the same situation and compare.** Same seed, same instrument, three windows a side.\n`scripts\u002Fcompare_windows.py`, then `reference\u002Fcompare.md`.\n\n### One turn of the loop\n\nThe complaint is *\"it hitches when the third pickup spawns in the hazard lane\"*. Booted into that\nsituation with a fixed seed, one window says:\n\n```\n[PERF] window=hazard-lane mode=editor-windowed frames=90 ft_med=8.31 ft_p95=12.90 ft_max=41.2\n       hitches=2 draw_med=412 tris_med=51200 gc_sum=184320 disc[pickup-pool]=214 missing=1 zero=0\n```\n\nRead it in that order. The median is fine, so the game is not uniformly heavy; `ft_p95` at 12.90\nagainst a median of 8.31 is unevenness. Draw calls and triangles are the same as the run before,\nso it is not a rendering cost at all. The last number is the finding: the pickup pool **threw away\n214 units of work** during those ninety frames while completing seventeen. The lane selector is\nflipping across a 1.3-unit lane boundary and cancelling the set it was building each time.\n\nA 0.2-unit dead band on that boundary, then the same window again: discarded 214 → 7, `ft_p95`\n12.90 → 9.10 ms, hitches 2 → 0, triangles unchanged. One parameter, two numbers that moved\ntogether, and a table that says why.\n\n---\n\n## Where the time goes\n\nA router, read against a window you already have. Depth for the first four rows is in\n`reference\u002Fwhere-the-time-goes.md`; the rest belong to the skill that owns the subsystem.\n\n| The window says | Go to |\n|---|---|\n| One large `ft_max` the first time something appears, never again | first-appearance section, then `unity-render-urp` |\n| `Memory\u002FGC Allocated In Frame` has a non-zero median; spikes on a period | GC section, then `unity-game-ui` for per-frame string building |\n| A spike lined up with a jump in a completed-work counter | synchronous-loads section, then `unity-audio` when it is a clip |\n| Triangles high while looking at distance, draw calls ordinary | distant-geometry section, then `unity-scene-optimisation` |\n| Discarded work climbing while scheduled climbs | `reference\u002Fwork-counters.md` |\n| Draw calls and SetPass calls high, triangles ordinary | `unity-scene-optimisation` · `unity-sprite-atlas` for 2D · `unity-game-ui` for canvases |\n| GPU frame time much larger than the CPU times | `unity-scene-optimisation` · `unity-render-urp` for post-processing |\n| Shadow caster count high | `unity-lighting` |\n| Pixel art shimmering as well as slow | `unity-2d-pixel-perfect` |\n| Slow only in the browser | `unity-web-release` |\n| Slow only on the device, fine in the editor | `unity-device-testing`, after a development-player window taken on it |\n| Frame time is fine and it still feels wrong | `unity-play-harness` |\n\n---\n\n## Discipline\n\n- **List, never recall.** Counter names are defined by the build, not by your memory of them. One\n  catalogue call is cheaper than one wrong name, and a wrong name reports zero rather than failing.\n- **A counter that reads zero on every frame of a window is an absence, not a measurement of\n  nothing.** The window separates the two: `missing[]` is \"not exposed on this build\",\n  `zeroAll[]` is \"exposed and flat\". Report both; never quietly average a zero into a result.\n- **A quiet machine, or do not measure.** Another Editor or a Blender job turned 1.6–2.4 ms into\n  4–10 ms in one project; check the process list first (`reference\u002Fframe-hygiene.md`).\n- **Same situation, same instrument, or do not compare.** Different run mode, graphics device,\n  screen size, vsync or warm-up makes two numbers that subtract cleanly and mean nothing.\n- **Median and worst, never just worst** — one spike and a constant vibration are different faults\n  — with p95 between them to say where the bad frames start.\n- **Warm up 30 frames by default.** The first frames after entering a situation measure starting,\n  not playing. Set warm-up to 0 deliberately when a first-appearance hitch *is* the subject, and\n  put a `-cold` suffix on that window's label so nobody compares it against a warmed one.\n- **Three windows a side before believing a 10% change.** Below that, one window each way is a\n  coin toss, and the comparison script will tell you when the spread makes the change unreadable.\n- **Delete the JSONL before every run.** It is appended to. A stale window compared against a\n  fresh one is the quietest wrong answer this skill can produce.\n- **Print the numbers the judgement rests on.** \"The hitch is gone\" ages badly; \"discarded 214 → 7,\n  p95 12.90 → 9.10 ms\" keeps saying the same thing after the next change.\n- **Check whether the candidate is faster because it does less.** Fewer things built is not the\n  same as things built more cheaply. Read the scheduled and completed counters before celebrating.\n- **Every report names its run mode and what it did not cover.** A report that omits the\n  instrument invites the reader to assume it was the shipped game.\n\n## Scope — what this skill does NOT do\n\n| Not here | Go to |\n|---|---|\n| Booting into a moment, a readable state line, a feel complaint with nothing to measure | `unity-play-harness` |\n| Driving Unity, exit codes, screenshots, proving a harness told the truth | `unity-debug` |\n| Installing editors, licences, CI invocation | `unity-cli` |\n| Atlasing sprites, decimating meshes | `unity-sprite-atlas` · `unity-3d-models` |\n| Configuring post-processing and renderer features | `unity-render-urp` |\n| Browser download size, headers and memory ceilings | `unity-web-release` |\n| Building, signing and reading a device artifact back | `unity-android-release` |\n| Stutters on the device, runs fine in the Editor — a window taken on the phone | `unity-device-testing` |\n| Audio memory, load types and mixer cost | `unity-audio` |\n| Deciding what the budget should be | `unity-game-brief` |\n| Where a character's frame windows fall: the budget ladder, the import check | `game-character-pipeline` |\n",{"name":5,"description":7},"main","b7459aa49fb881c5abb764223674ec8f397a02b1","skills\u002Funity-profiling",0,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills","https:\u002F\u002Fgithub.com\u002FtipoLi5890","https:\u002F\u002Ftipoli5890.github.io\u002F","2026-11-10 09:16:41",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add tipoLi5890\u002Funity-dev-skills --skill unity-profiling",{"tool":38,"command":39},"bunx","bunx skills add tipoLi5890\u002Funity-dev-skills --skill unity-profiling",{"tool":41,"command":42},"pnpm","pnpm dlx skills add tipoLi5890\u002Funity-dev-skills --skill unity-profiling","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002FSKILL.md",true,[46,52,56,60,64,68,72,76,80,84,88,94,98,102,106,110,114,118,124,128],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"reference\u002Fcompare.md","REFERENCE","参考文档",6598,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fcompare.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"reference\u002Fcounters.md",7368,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fcounters.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"reference\u002Fframe-hygiene.md",3873,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fframe-hygiene.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"reference\u002Fframe-windows.md",8184,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fframe-windows.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"reference\u002Fgc-per-phase.md",3320,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fgc-per-phase.md",{"path":69,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":70,"sourceHtmlUrl":71},"reference\u002Fperf-tests.md",11355,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fperf-tests.md",{"path":73,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":74,"sourceHtmlUrl":75},"reference\u002Fraw-capture.md",5963,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fraw-capture.md",{"path":77,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":78,"sourceHtmlUrl":79},"reference\u002Fruntime-asset-leaks.md",2439,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fruntime-asset-leaks.md",{"path":81,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":82,"sourceHtmlUrl":83},"reference\u002Fwhere-the-time-goes.md",7733,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fwhere-the-time-goes.md",{"path":85,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":86,"sourceHtmlUrl":87},"reference\u002Fwork-counters.md",6200,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Freference\u002Fwork-counters.md",{"path":89,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":92,"sourceHtmlUrl":93},"resources\u002FCounterCatalog.cs","OTHER","其他",6869,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FCounterCatalog.cs",{"path":95,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":96,"sourceHtmlUrl":97},"resources\u002FFrameSampler.cs",30875,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FFrameSampler.cs",{"path":99,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":100,"sourceHtmlUrl":101},"resources\u002FGcAllocProbe.cs",6322,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FGcAllocProbe.cs",{"path":103,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":104,"sourceHtmlUrl":105},"resources\u002FGcAllocProbeOpener.cs",529,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FGcAllocProbeOpener.cs",{"path":107,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":108,"sourceHtmlUrl":109},"resources\u002FOwnedAssets.cs",2202,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FOwnedAssets.cs",{"path":111,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":112,"sourceHtmlUrl":113},"resources\u002FProfilerRawCapture.cs",10823,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FProfilerRawCapture.cs",{"path":115,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":116,"sourceHtmlUrl":117},"resources\u002FWorkCounters.cs",6957,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fresources\u002FWorkCounters.cs",{"path":119,"resourceType":120,"resourceTypeDesc":121,"sizeBytes":122,"sourceHtmlUrl":123},"scripts\u002Fcompare_windows.py","SCRIPT","脚本",19047,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Fscripts\u002Fcompare_windows.py",{"path":125,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":126,"sourceHtmlUrl":127},"templates\u002FAssetLeakSoakTest.cs",5096,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Ftemplates\u002FAssetLeakSoakTest.cs",{"path":129,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":130,"sourceHtmlUrl":131},"templates\u002FPerfRegressionTest.cs",15446,"https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-profiling\u002Ftemplates\u002FPerfRegressionTest.cs",[133,140,147,154,161,168],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":19},"01M7CMNYXQNW14ST64EYEEZ5V5","unity-web-release","Ship a WebGL\u002FWebGPU build: shrink the download, Brotli server headers, Safari and iPhone memory.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-web-release\u002FSKILL.md",[],[],{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":19},"01M7CMNTHD1EA70CCJ2F9FN3HT","unity-vivox","Vivox voice and text chat: channels, positional voice, mute and push-to-talk, mic permission.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-vivox\u002FSKILL.md",[],[],{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":19},"01M7CMNPHPRQZ7K1FES2953DRQ","unity-urp-migration","Migrate a Built-in project to URP, or fix a half-moved one: pink materials, PPv2, old shaders.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-urp-migration\u002FSKILL.md",[],[],{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":19},"01M7CMNJ756DK4V0D50HMAZYQW","unity-ui-ugui","Canvas UI (uGUI): CanvasScaler, RectTransform anchors, Layout Groups, ScrollView, UI built from C#.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-ui-ugui\u002FSKILL.md",[],[],{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":167,"publishedAt":19},"01M7CMNDESZN79EK5FRDXCA2WA","unity-ui-toolkit","UI Toolkit (UXML, USS, UIDocument): flex layout, binding, custom elements, editor windows.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-ui-toolkit\u002FSKILL.md",[],[],{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":19},"01M7CMN8GBV07K0YCY07BJ6A1W","unity-ugs","Unity Gaming Services API: which package, init and sign-in order, Cloud Save, Cloud Code, Remote Config.","https:\u002F\u002Fgithub.com\u002FtipoLi5890\u002Funity-dev-skills\u002Fblob\u002Fmain\u002Fskills\u002Funity-ugs\u002FSKILL.md",[],[]]