kali20gakki/msAgent / skills/profiler/msagent-profiler-breakdown
msagent-profiler-breakdown
VerifiedDB格式Profiling数据场景化性能拆解:自动完成场景确认(框架/任务类型判定)、阶段拆解与单次执行边界拆解,并给出报告。Use when 需要对昇腾 Profiling DB 做阶段拆解、step 归因或生成性能报告, or when the user says "帮我按任务阶段拆解这个Profiling数据", "多层次性能耗时拆解", "分析这个 vllm/sglang/verl 数据,主要耗时都花费在哪里了"。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 8.0 KB |
| references/breakdown-schema.md | markdown | 4.1 KB |
| references/check-rules.md | markdown | 1.5 KB |
| references/db-usage.md | markdown | 5.8 KB |
| references/frameworks/README.md | markdown | 5.5 KB |
| references/frameworks/verl.md | markdown | 11.9 KB |
| references/frameworks/vllm-instrumentation.md | markdown | 13.6 KB |
| references/frameworks/vllm-profiler-data.md | markdown | 4.6 KB |
| references/frameworks/vllm.md | markdown | 5.6 KB |
| references/report-template.md | markdown | 5.0 KB |
| references/scenario-taxonomy.md | markdown | 5.7 KB |
| resources/scenarios/inference/sglang.json | config | 549 B |
| resources/scenarios/inference/vllm.json | config | 912 B |
| resources/scenarios/rl/slime.json | config | 468 B |
| resources/scenarios/rl/verl.json | config | 1.4 KB |
| resources/scenarios/training/fsdp.json | config | 303 B |
| resources/scenarios/training/megatron.json | config | 556 B |
| resources/scenarios/training/mindspeed.json | config | 318 B |
| scripts/compare_runs.py | script | 12.4 KB |
| scripts/db_query.py | script | 12.6 KB |
| scripts/decompose.py | script | 47.7 KB |
references/check-rules.md
View on GitHub# 校验与判定规则(守恒 / 阈值 / 置信度)
> 消费脚本:`msagent-profiler-breakdown/scripts/decompose.py`(阶段/step 拆解 + 报告生成)
> 作用:把「结果是否正确、多慢算异常、可不可信」的判据集中为可配置规则。
## 1. 守恒自检阈值
- 缺口定义:`conservation_gap = parent.wall_time − Σ children.self_time`。
- 分类:重叠(overlap)/ 未归因(custom/unknown)/ rounding。
- 告警阈值(默认):`gap > parent.wall_time × 5%` → 显式告警「归属不完整」。
- 范围:端到端\~层内各做一次,逐级上报,缺口值写入 `confidence`。
## 2. 异常/漂移阈值(可配置)
- 最差 step/层 outlier:`duration > P50 + k·σ`(k 默认 3)或 `> P99`。
- 漂移:`CV = σ/μ > 0.20` 告警(同卡跨 step)。
- 跨 run:均值 ± σ 区间对比。
- 稳态口径:剔除首 step 热身/编译噪声 + ±1 tail。
## 3. 置信度分级
| 级别 | 依据 | 呈现 |
| ------ | ------------------------------------ | --------- |
| high | 框架打点 / record\_function / STEP\_TIME | 正常呈现 |
| medium | 锚点 + 层数校验 / 序列对齐 | 标注方法 |
| low | C 策略推测 / custom/unknown / 守恒缺口超阈值 | 逐条 + 提示复核 |
## 4. 配置方式
- 阈值集中于此文件;引擎读取,允许用户按场景覆盖(如放宽/收紧 outlier 的 k 与漂移 CV)。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |