SkillsAdd

Source

bbuf/ai-infra-auto-driven-skills

23 skills · 847 combined installs

Skills from this source

#
Skill
Source
8W Activity
Installs
1
sglang-prod-incident-triage Replay-first debug flow for SGLang serving problems. Use when a live or recent server shows health-check failures, la…
bbuf/ai-infra-auto-driven-skills
89
2
llm-serving-auto-benchmark Framework-independent LLM serving benchmark skill for comparing SGLang, vLLM, TensorRT-LLM, TokenSpeed, or another se…
bbuf/ai-infra-auto-driven-skills
85
3
llm-torch-profiler-analysis Unified LLM torch-profiler triage skill for `sglang`, `vllm`, `TensorRT-LLM`, and `TokenSpeed`. Use it to inspect an …
bbuf/ai-infra-auto-driven-skills
80
4
model-architecture-diagram Return public original model architecture diagrams for user-specified LLM, VLM, MoE, diffusion, OCR, and SGLang/sgl-c…
bbuf/ai-infra-auto-driven-skills
74
5
llm-serving-capacity-planner Parse SGLang/vLLM startup logs to explain GPU memory use and request capacity. Use for KV cache budget, mem-fraction-…
bbuf/ai-infra-auto-driven-skills
53
6
sglang-sota-humanize-loop Run an autonomous Humanize-governed SGLang SOTA performance loop for one LLM model: first perform a fixed fair SGLang…
bbuf/ai-infra-auto-driven-skills
53
7
llm-pipeline-analysis Inspect LLM torch profiler traces at forward-pass, layer, and kernel level. Use when you need layer timings, anchor-k…
bbuf/ai-infra-auto-driven-skills
52
8
model-pr-history-knowledge Use when an SGLang, vLLM, TensorRT-LLM, or TokenSpeed serving/model optimization task needs prior model-family PR evi…
bbuf/ai-infra-auto-driven-skills
51
9
vllm-sota-humanize-loop Run an autonomous Humanize-governed vLLM SOTA performance loop for one LLM model: first perform the fixed fair vLLM/S…
bbuf/ai-infra-auto-driven-skills
47
10
model-compute-simulation Build an operator-level compute template for an LLM and estimate FLOPs/MFU for a serving shape. Use when you need ten…
bbuf/ai-infra-auto-driven-skills
46
11
sglang-humanize-review Perform SGLang code review in the style of human maintainers by consulting the full non-agent PR review episode corpu…
bbuf/ai-infra-auto-driven-skills
45
12
h100 SSH into host `h100_sglang`, enter Docker container `sglang_bbuf`, work in `/sgl-workspace/sglang`, and use the ready…
bbuf/ai-infra-auto-driven-skills
36
13
h100-sglang-diffusion SSH into host `h100_sglang`, enter Docker container `sglang_bbuf`, work in `/data/bbuf/repos/sglang`, and use the rea…
bbuf/ai-infra-auto-driven-skills
36
14
model-pr-diff-dossier Use when creating or revising model PR optimization history documents for SGLang, vLLM, or another serving framework …
bbuf/ai-infra-auto-driven-skills
22
15
sglang-sota-performance End-to-end SGLang SOTA performance workflow. Use when a user names an LLM model and wants SGLang to match or beat the…
bbuf/ai-infra-auto-driven-skills
17
16
sglang-kimi-k2-k25-optimization PR-backed and current-main optimization manual for `moonshotai/Kimi-K2*` and `moonshotai/Kimi-K2.5*` in SGLang. Use w…
bbuf/ai-infra-auto-driven-skills
15
17
sglang-minimax-m2-series-optimization PR-backed and current-main optimization manual for the `MiniMaxAI/MiniMax-M2` series, including M2, M2.1, M2.5, M2.7,…
bbuf/ai-infra-auto-driven-skills
15
18
sglang-torch-profiler-analysis Compact SGLang torch-profiler triage skill. Use when Codex should inspect an existing `trace.json(.gz)` or profile di…
bbuf/ai-infra-auto-driven-skills
15
19
gpu-kernel-ako4all Use when developing, optimizing, debugging, or porting AI-infra GPU kernels through an AKO4ALL-centered loop, includi…
bbuf/ai-infra-auto-driven-skills
11
20
sglang-model-day0-support Build or audit an evidence-driven SGLang Day-0 support program for a new LLM, VLM, MoE, hybrid-attention, or speculat…
bbuf/ai-infra-auto-driven-skills
2
21
sglang-deepseek-v3-r1-optimization PR-backed and current-main optimization manual for DeepSeek V3 and DeepSeek R1 in SGLang. Use when Codex needs to rec…
bbuf/ai-infra-auto-driven-skills
1
22
sglang-deepseek-v31-optimization PR-backed and current-main optimization manual for DeepSeek V3.1 and DeepSeek-V3.1-Terminus in SGLang. Use when Codex…
bbuf/ai-infra-auto-driven-skills
1
23
sglang-deepseek-v32-optimization PR-backed and current-main optimization manual for DeepSeek V3.2, V3.2-Exp, V3.2-Speciale, NVFP4, and MXFP4 in SGLang…
bbuf/ai-infra-auto-driven-skills
1