tensorrt-llm

75 / 100 · Review before installing

Check before installing. Some of what it can do is expected: a coding skill usually runs build and test commands.

By raphaelmansuy/edgecrab (GitHub) · coding · 94 stars · scanned at commit ed33307

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.

Works with any agent that supports SKILL.md skills.

What this skill can do

Runs commands on your computer

The author does not mention this. Expected for a coding skill.

Found in 2 places.

skills/mlops/inference/tensorrt-llm/SKILL.md, line 43docker pull nvidia/tensorrt_llm:latest
skills/mlops/inference/tensorrt-llm/SKILL.md, line 78trtllm-serve meta-llama/Meta-Llama-3-8B \

Downloads and installs other software

The author does not mention this.

Found in 2 places.

skills/mlops/inference/tensorrt-llm/SKILL.md, line 46pip install tensorrt_llm==1.2.0rc3
skills/mlops/inference/tensorrt-llm/references/optimization.md, line 208pip install tensorrt_llm[benchmark]

Connects to the internet

The author does not mention this.

Found in 12 places.

skills/mlops/inference/tensorrt-llm/SKILL.md, line 84curl -X POST http://localhost:8000/v1/chat/completions \
skills/mlops/inference/tensorrt-llm/references/multi-gpu.md, line 232curl http://localhost:8000/metrics
skills/mlops/inference/tensorrt-llm/references/serving.md, line 74curl -X POST http://localhost:8000/v1/chat/completions \

Show 9 more

skills/mlops/inference/tensorrt-llm/references/serving.md, line 114curl -X POST http://localhost:8000/v1/chat/completions \
skills/mlops/inference/tensorrt-llm/references/serving.md, line 137curl -X POST http://localhost:8000/v1/completions \
skills/mlops/inference/tensorrt-llm/references/serving.md, line 161curl http://localhost:9090/metrics
skills/mlops/inference/tensorrt-llm/references/serving.md, line 177curl http://localhost:8000/health/ready
skills/mlops/inference/tensorrt-llm/references/serving.md, line 180curl http://localhost:8000/health/live
skills/mlops/inference/tensorrt-llm/references/serving.md, line 183curl http://localhost:8000/v1/models
skills/mlops/inference/tensorrt-llm/references/serving.md, line 395curl -X POST http://localhost:8000/v1/chat/completions \
skills/mlops/inference/tensorrt-llm/references/serving.md, line 431curl http://localhost:9090/metrics | grep queue_size
skills/mlops/inference/tensorrt-llm/references/serving.md, line 434curl http://localhost:9090/metrics | grep active_requests

Safer alternatives

Low-risk skills in the same category (coding) that look related. Check that one does what you need.

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'tensorrt-llm' ] || [ -L ~/.agents/skills/'tensorrt-llm' ]; then echo 'A skill folder named tensorrt-llm already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/raphaelmansuy/edgecrab' "$d" && git -C "$d" checkout --quiet 'ed3330703bc91f498062451136b89592c9ee098a' -- 'skills/mlops/inference/tensorrt-llm' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/mlops/inference/tensorrt-llm' ~/.agents/skills/'tensorrt-llm' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'tensorrt-llm' ~/.claude/skills/'tensorrt-llm' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for raphaelmansuy/edgecrab shows the score of its lowest-scoring skill.

SkillTrust badge for raphaelmansuy/edgecrab

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/raphaelmansuy/edgecrab.svg)](https://zalavijaysinh.github.io/skilltrust/repos/raphaelmansuy/edgecrab/)

Score history

Date	Version	Score
2026-11-14	`ed33307`	75

Also found in

Identical copies of this skill exist at:

Tommy-yw/RunbookHermes/optional-skills/mlops/tensorrt-llm