New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Clean Startup Log

by sgl-project in sgl-project/sglang

Audit SGLang startup logs, save evidence, and propose cleanup for user review.

Apache-2.0Auto-check passedAI & LLM Engineering

Install Clean Startup Log

skills CLI
$ npx skills add sgl-project/sglang --skill clean-startup-log -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install sgl-project/sglang clean-startup-log --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/sgl-project/sglang.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/clean-startup-log .claude/skills/clean-startup-log && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
clean-startup-log
GitHub stars
37k
Token cost
~1.5k tokens
SKILL.md length
755 words
Files
2 (incl. references)
Skills in repo
31
Repo updated
First seen
Licence
Apache-2.0

At a glance

Audit SGLang startup logs, save evidence, and propose cleanup for user review.

  • Works in 4 steps: Check the checkout, free GPUs, and ports… → Create a unique directory with mktemp -d… → Wait for The server is fired up and… → …
  • AI & LLM Engineering work in your project
  • SKILL.md covers Default runs, Capture logs, Investigate efficiently and Accepted output, plus 2 more sections
  • Calls uv

What it does

Clean Startup Log is an agent skill from sgl-project/sglang. Audit SGLang startup logs, save evidence, and propose cleanup for user review. With no arguments, run Qwen3-8B at TP1 and TP2 plus gpt-oss-20b at TP1.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/noise-sources.md`).

It sits in AI & LLM Engineering. It works with SGLang and Qwen. The repository describes itself as: SGLang is a high-performance serving framework for large language models and multimodal models. The licence is Apache-2.0.

When your agent uses it

  • AI & LLM Engineering work in your project

Example prompts

  • “/clean-startup-log”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Check the checkout, free GPUs, and ports once. Use the requested command when
  2. Create a unique directory with mktemp -d /tmp/sglang-startup-audit-XXXXXX.
  3. Wait for The server is fired up and ready to roll!, then stop that server and
  4. Preserve the user's logging configuration, including NCCL_DEBUG. If NCCL

What it can do on your machine

Read from SKILL.md and the folder at commit b7b2975. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • uv

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Clean Startup Log loads about 1.5k tokens when it runs, and up to ~2.6k if it reads all its reference files. Until then it costs about 42 tokens; SKILL.md has 755 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~42
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from sgl-project/sglang at commit b7b2975, republished under its Apache-2.0 licence (© sgl-project). 755 words, ~1,495 tokens.

Download SKILL.mdSave it as .claude/skills/clean-startup-log/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
clean-startup-log
description
Audit SGLang startup logs, save evidence, and propose cleanup for user review. With no arguments, run Qwen3-8B at TP1 and TP2 plus gpt-oss-20b at TP1.
disable-model-invocation
true

Audit SGLang Startup Logs

The default outcome is saved logs and a findings report. Apply runtime changes only after the user selects them. A request to edit this skill does not itself launch servers.

Default runs

A bare $clean-startup-log invocation runs these cases sequentially without asking for commands. Explicit commands, models, or TP sizes replace this matrix.

Case / log filenameCommand
qwen3-8b-tp1.loguv run sglang serve --model-path Qwen/Qwen3-8B
qwen3-8b-tp2.loguv run sglang serve --model-path Qwen/Qwen3-8B --tp 2
gpt-oss-20b-tp1.loguv run sglang serve --model-path openai/gpt-oss-20b

These cover dense, tensor-parallel, and MoE/hybrid sliding-window attention startup. Reuse complete captures from the current audit when code and environment have not changed.

Capture logs

  1. Check the checkout, free GPUs, and ports once. Use the requested command when resources are free; otherwise select free GPUs with CUDA_VISIBLE_DEVICES and an unused --port, recording the adjustments. Leave existing servers alone.
  2. Create a unique directory with mktemp -d /tmp/sglang-startup-audit-XXXXXX. Save raw stdout and stderr together in a separate log for each case, for example with set -o pipefail and COMMAND 2>&1 | tee LOG_PATH. Record commands, GPU IDs, ports, commit, relevant overrides, and readiness status.
  3. Wait for The server is fired up and ready to roll!, then stop that server and its workers before the next case. First runs can spend many minutes downloading weights or compiling FlashInfer kernels; check download/compiler activity before treating a quiet log as a hang. Preserve partial logs for failed or stalled starts and report the last stage. Continue independent cases when possible.
  4. Preserve the user's logging configuration, including NCCL_DEBUG. If NCCL verbosity needs explaining, inspect relevant shell settings, NCCL_CONF_FILE, and /etc/nccl.conf. Do not override intentional diagnostics or recommend NCCL_DEBUG=WARN solely because the output is long. Avoid full environment dumps.

Investigate efficiently

  • Scan for deprecations, duplicate handler output, unrelated import failures, unformatted prints, and unexpected warnings. Read representative excerpts and counts instead of repeatedly dumping server_args, progress redraws, or NCCL diagnostics. Normalize carriage returns for analysis only; preserve raw logs.
  • Trace each candidate to its actual emitter with focused rg searches. Inspect its log level: SGLang's formatter may omit severity. Group shared signatures across cases and distinguish handler duplication from separate GPU/process calls.
  • Repetition, WARNING severity, or a different third-party format alone does not establish a cleanup need. Consider whether the message explains configuration, progress, resource use, or an operational limitation.
  • Consult noise-source hints only for a matching signature or an unresolved emitter. Verify current code rather than trusting historical line numbers, fix status, or assumptions about unrelated models.
Show full SKILL.md (337 more words)Show less

Accepted output

Preserve these reviewed messages unless the user requests a different policy:

  • NCCL diagnostics enabled by the user's environment or host configuration.
  • NUMA permission warnings, including one check per GPU in TP runs.
  • GPT-OSS MXFP4 backend-selection warnings and default page-size selection warnings.
  • Init Unified Radix Cache. Components: ... Tree Core: ..., tree-cache summaries, SWA allocation details, and per-rank memory/timing records.
  • Useful progress bars, warmup HTTP access logs, uv synchronization messages, isolated NCCL/Gloo startup lines, and one timestamped HF authentication warning.

These can appear in a clean startup log. Do not repeatedly propose the declined NUMA deduplication, backend/page-size level changes, or NCCL verbosity override. Keep real operational warnings visible: for example, a Harmony vocabulary failure can disable /v1/responses even when server readiness and /generate succeed.

Report before changing code

Return a compact run table with readiness status and clickable raw-log links. For each actual cleanup candidate, give an exact representative message, affected cases/counts, source file/function, and specific proposed behavior. Distinguish confirmed findings from suspicions and operational failures from logging noise.

If there are no actionable cleanup findings, say the logs are clean and no further cleanup is needed. Otherwise, ask which numbered changes to adopt and wait for the user's selections before editing runtime code or preparing patches. Honor existing approvals and declined items without asking again.

Apply selected changes

  • Batch compatible approved edits, then verify affected cases once. Repeat a startup only for a new change, failure, or unresolved concern; do not relaunch after every one-line edit. Save verification logs separately from baselines.
  • Preserve useful warnings and application handlers. HF can warn during early CLI model detection before configure_logger(), and spawned processes have independent logger state. Keep configure_hf_hub_logger() in both suppress_noisy_warnings() and configure_logger(); make repeated setup safe.
  • Keep the legacy compiled-kernel cache migration notice at DEBUG. Avoid broad library-level suppression or fd redirection for a narrow logging problem.
  • Run relevant formatting and focused existing checks. Add tests only when they verify meaningful behavior, not a log-level spelling. Report changes and verification; create branches, commits, and PRs when requested.

© sgl-project, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in .agents/skills/clean-startup-log of sgl-project/sglang.

  • SKILL.md
  • references/noise-sources.md

Open the folder on GitHubat commit b7b2975

Compare with similar skills

Clean Startup Log next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Clean Startup Log compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Clean Startup Log this skillsgl-project/sglang37k—~1.5kAutomated safety check: PassApache-2.0
LLM Pipeline Profiler AnalysisBBuf/AI-Infra-Auto-Driven-SKILLS911—~3.9kAutomated safety check: PassNone
Hyperloom Remote Mn Qwen3 30bAMD-AGI/Hyperloom217—~1.8kAutomated safety check: PassCustom licence
Model Architecture Diagram FinderBBuf/AI-Infra-Auto-Driven-SKILLS911—~1.2kAutomated safety check: PassNone
slime RL Post-TrainingOrchestra-Research/AI-Research-SKILLs13k5 repos~2.8kAutomated safety check: PassMIT
SageMaker Serving Image Selectionhuggingface/skills11k1 repos~4.6kAutomated safety check: PassApache-2.0

Similar skills

  • LLM Pipeline Profiler Analysis

    BBuf/AI-Infra-Auto-Driven-SKILLS

    Breaks LLM torch profiler traces down by forward pass, layer and kernel, with timing tables and Perfetto time ranges for the layers you want to inspect.

    911 GitHub stars~3.9k tokensUpdated 3 days ago
    AI & LLM EngineeringAuto-check passed

  • Hyperloom Remote Mn Qwen3 30b

    AMD-AGI/Hyperloom

    Run a 4-hour multi-node Hyperloom Qwen3-30B-A3B optimization (Infera PD-disaggregated or RayJob aggregated) with --nodes 2 and sglang MoE tuning on MI325X.

    217 GitHub stars~1.8k tokensUpdated today
    AI & LLM EngineeringAuto-check passed

  • Model Architecture Diagram Finder

    BBuf/AI-Infra-Auto-Driven-SKILLS

    Looks up public original architecture diagrams for named LLM, vision-language, MoE, diffusion and OCR models and returns the image with its source attribution.

    911 GitHub stars~1.2k tokensUpdated 3 days ago
    AI & LLM EngineeringAuto-check passed

  • slime RL Post-Training

    Orchestra-Research/AI-Research-SKILLs

    Guides reinforcement-learning post-training of LLMs with slime, which pairs Megatron-LM training with SGLang rollouts, including GRPO runs on GLM, Qwen3 and Llama 3 models.

    13k GitHub starsUsed in 5 repos~2.8k tokens
    AI & LLM EngineeringAuto-check passed

  • SageMaker Serving Image Selection

    huggingface/skills

    Official

    Chooses the right serving container and current image URI for deploying a Hugging Face model to a SageMaker endpoint, preferring Hugging Face images over generic ones.

    11k GitHub starsUsed in 1 repo~4.6k tokens
    AI & LLM EngineeringAuto-check passed

  • Fix Art Issues

    OpenPipe/ART

    Fix a GitHub issue on OpenPipe/ART and open a PR. An agent skill from OpenPipe/ART.

    11k GitHub stars~840 tokensUpdated yesterday
    AI & LLM EngineeringAuto-check: notes

More from sgl-project/sglang

All 31 skills in this repo

  • Sglang Prod Incident Triage

    sgl-project/sglang

    Replay-first debug flow for SGLang serving problems. An agent skill from sgl-project/sglang.

    37k GitHub starsUsed in 3 repos~2.1k tokens
    Auto-check passed

  • LLM Torch Profiler Analysis

    sgl-project/sglang

    Unified LLM torch-profiler triage skill for sglang, vllm, TensorRT-LLM, and TokenSpeed.

    37k GitHub starsUsed in 2 repos~6.4k tokens
    Auto-check passed

  • Babysit PR To Pass CI

    sgl-project/sglang

    Start and persistently pursue a goal to babysit an SGLang pull request until selected GitHub Actions workflows pass on the latest PR head.

    37k GitHub starsUsed in 2 repos~3k tokens
    Auto-check passed

  • Compute Mamba Ratio

    sgl-project/sglang

    Compute the optimal --mamba-full-memory-ratio (or --max-mamba-cache-size pin) for a hybrid attention + linear-attention (Mamba / GDN / KDA) model's two serving memory pools, from the workload and…

    37k GitHub starsUsed in 2 repos~2.9k tokens
    Auto-check passed

  • Debug Distributed Hang

    sgl-project/sglang

    Debug hanging issues in SGLang distributed inference (TP/PP/DP/EP).

    37k GitHub starsUsed in 2 repos~2.4k tokens
    Auto-check passed

  • Env Var Conventions

    sgl-project/sglang

    Conventions for SGLang environment variables — where to define, how to access, how to name, and how to deprecate.

    37k GitHub starsUsed in 2 repos~2.9k tokens
    Auto-check passed

Works with

Categories

Questions about Clean Startup Log

What does Clean Startup Log do?

Audit SGLang startup logs, save evidence, and propose cleanup for user review. Clean Startup Log is an agent skill from sgl-project/sglang. Audit SGLang startup logs, save evidence, and propose cleanup for user review.

When should I use Clean Startup Log?

Clean Startup Log fits situations like: AI & LLM Engineering work in your project.

How do I install Clean Startup Log in Claude Code?

Run `npx skills add sgl-project/sglang --skill clean-startup-log -a claude-code`. Or copy the skill folder (.agents/skills/clean-startup-log in sgl-project/sglang) into .claude/skills/clean-startup-log in your project. Claude Code loads it when a task matches its description.

How do I install Clean Startup Log in Codex?

Run `npx skills add sgl-project/sglang --skill clean-startup-log -a codex`. Or copy the skill folder (.agents/skills/clean-startup-log in sgl-project/sglang) into .agents/skills/clean-startup-log in your project. Codex loads it when a task matches its description.

Can I use Clean Startup Log in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add sgl-project/sglang --skill clean-startup-log -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/clean-startup-log, .gemini/skills/clean-startup-log, .github/skills/clean-startup-log and .opencode/skills/clean-startup-log in your project.

What does Clean Startup Log need to run?

Going by SKILL.md and its folder, Clean Startup Log needs the command-line tools its instructions call (uv).

Does Clean Startup Log access the network?

SKILL.md contains no URLs. Its commands use uv, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Clean Startup Log safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Clean Startup Log use?

Clean Startup Log is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Clean Startup Log use?

About 1.5k tokens (SKILL.md is roughly 6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.1k tokens, read only when the agent opens those files.

What are the alternatives to Clean Startup Log?

Skills that share tags, products or a category with Clean Startup Log: LLM Pipeline Profiler Analysis (BBuf/AI-Infra-Auto-Driven-SKILLS, 911 stars), Hyperloom Remote Mn Qwen3 30b (AMD-AGI/Hyperloom, 217 stars), Model Architecture Diagram Finder (BBuf/AI-Infra-Auto-Driven-SKILLS, 911 stars) and slime RL Post-Training (Orchestra-Research/AI-Research-SKILLs, 13k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Clean Startup Log?

sgl-project (a GitHub organization) maintains it in sgl-project/sglang, which has 36,851 GitHub stars. The repository holds 31 skills in this directory. The repository was last updated on October 8, 2026.

Source: sgl-project/sglang on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.