tensorrt-llm

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.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured, actionable, and cleanly offloads detail to real reference files, scoring high on actionability and progressive disclosure. It loses points on conciseness from version pins and marketing claims, and lacks explicit validation checkpoints in its workflows.

Suggestions

Move or remove brittle version pins (tensorrt_llm==1.2.0rc3, CUDA 13.0.0) from the main body into a dedicated compatibility/deprecated section to keep the overview evergreen.

Trim marketing phrasing like 'state-of-the-art performance' and unverified '100x faster' claims, keeping only concrete, verifiable benchmark numbers.

Add explicit verification steps (e.g., a health check after trtllm-serve start, or confirming GPU visibility) to the serving quick-start workflow.

Dimension	Reasoning	Score
Conciseness	Mostly lean with executable code and bullet lists, but includes version pins (tensorrt_llm==1.2.0rc3, CUDA 13.0.0) and marketing-style claims ('state-of-the-art performance', '100x faster') that add tokens outside a deprecated/old-patterns section.	2 / 3
Actionability	Fully executable, copy-paste-ready code and commands — LLM/SamplingParams Python API, docker/pip install, trtllm-serve invocation, and a curl client request.	3 / 3
Workflow Clarity	Sequences are present (install → serve → client request) but there are no explicit validation or verification checkpoints; for a feature/reference skill this is acceptable but not fully instrumented.	2 / 3
Progressive Disclosure	Clear overview body with well-signaled, one-level-deep references to references/optimization.md, references/multi-gpu.md, and references/serving.md, all of which exist as real files.	3 / 3
	Total	10 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states capabilities, hardware context, and explicit use-when triggers in third person. It is concise yet comprehensive and unlikely to conflict with other skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and capabilities — 'Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency' plus 'quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.'	3 / 3
Completeness	Explicitly answers both what ('Optimizes LLM inference...') and when ('Use for production deployment...', 'when you need 10-100x faster inference', 'or for serving models with...').	3 / 3
Trigger Term Quality	Natural terms a user would say are well covered — 'LLM inference', 'NVIDIA GPUs (A100/H100)', 'quantization', 'production deployment', 'faster inference', 'multi-GPU scaling'.	3 / 3
Distinctiveness Conflict Risk	Clear hardware-specific niche (NVIDIA TensorRT production inference) with distinct triggers unlikely to overlap with generic serving skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/ml-inference/tensorrt-llm/SKILL.md
Commit: 52845c3

Reviewed: about 14 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.