SGLang is a framework for running inference for large language models and multimodal models, meaning it processes inputs to produce model outputs such as text or other media. It is used to serve and accelerate open AI models and related workloads.
Borrowing it
Nothing to install: this file belongs to sgl-project/sglang. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sgl-project/sglang/main/.agents/skills/clean-startup-log/SKILL.mdgit clone --depth 1 https://github.com/sgl-project/sglangWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sgl-project/sglang/clean-startup-log)<a href="https://agentmods.dev/skills/sgl-project/sglang/clean-startup-log"><img src="https://agentmods.dev/badge/skills/sgl-project/sglang/clean-startup-log/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sgl-project/sglang/clean-startup-log"><img src="https://agentmods.dev/badge/skills/sgl-project/sglang/clean-startup-log.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Snyk pass
- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 282 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00051 | $0.01306 |
| Opus 5.5 | $0.00020 | $0.00522 |
| Sonnet 5.5 | $0.00010 | $0.00261 |
| Haiku 4.5 | $0.00005 | $0.00131 |
Grade A, and why
clean-startup-log scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
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 filename | Command |
|---|---|
qwen3-8b-tp1.log |
uv run sglang serve --model-path Qwen/Qwen3-8B |
qwen3-8b-tp2.log |
uv run sglang serve --model-path Qwen/Qwen3-8B --tp 2 |
gpt-oss-20b-tp1.log |
uv 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
- Check the checkout, free GPUs, and ports once. Use the requested command when
resources are free; otherwise select free GPUs with
CUDA_VISIBLE_DEVICESand an unused--port, recording the adjustments. Leave existing servers alone. - 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 withset -o pipefailandCOMMAND 2>&1 | tee LOG_PATH. Record commands, GPU IDs, ports, commit, relevant overrides, and readiness status. - 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. - 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 recommendNCCL_DEBUG=WARNsolely because the output is long. Avoid full environment dumps.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · -177 lines · +6 tokens per session a173099a353e
- 2mo ago First seen · 283 lines · 45 tokens per session scan A bead524743d1
clean-startup-log is a skill published in the GitHub repository sgl-project/sglang (38,324 stars, last pushed today), licensed Apache-2.0. It adds 51 tokens to every session and 1,306 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
perf-analyze
Launch and operate this repo's perf-analyze workflow, which DIAGNOSES a TensorRT-LLM serving deployment without applying changes — benchmark at one concurrency or a Pareto curve of them (tok/s/user vs tok/s/gpu), analytical SOL projection on by default (via the internal-perf-sol-analysis skill), nsys + ncu per-kernel…
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
python-debugpy
Debug Python: pdb REPL + debugpy remote (DAP).
resolve-handoff
Emit the complete Resolve handoff with exact modes, outcomes, test evidence, and publication state.
iterate-implementation
Only use when the user explicitly invokes this skill by name.
debug-with-langwatch
Root-cause production errors and misbehaving agent runs with LangWatch. Finds errored traces, inspects spans, checks monitor and evaluator scores, then narrows to a root cause. Use when something is failing or misbehaving in production (errors, bad answers, latency spikes).