tensorpool-gpu-cloud

On-demand GPU clusters and training jobs with git-style interface. Use when you need multi-node GPU clusters (B200, H200, H100), persistent NFS storage, or batch training jobs with the TensorPool CLI.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

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

A highly actionable, well-structured CLI reference with concrete commands and clear workflows, weakened mainly by repeated credential/pricing content and the absence of any bundle-file split for a skill this size.

Suggestions

Consolidate the TENSORPOOL_KEY credential check and the per-GPU pricing into a single section each instead of repeating them across Credential Setup, Authentication, Troubleshooting, Key features, Pricing, and Cost Awareness.

Add an inline validation checkpoint to each workflow (e.g. 'verify cluster status is RUNNING before `tp ssh`', 'confirm job is Completed before `tp job pull`') to make destructive/batch sequences self-checking.

Consider moving the full Troubleshooting catalog and detailed command tables into a reference bundle file linked from SKILL.md to reduce the monolithic body length.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of over-explanation of known concepts, but credential-check instructions and pricing figures are each repeated across three separate sections and could be consolidated.	3 / 5
Actionability	Provides fully executable, copy-paste-ready `tp` commands, a complete `tp.config.toml` example, and a concrete SLURM/torchrun distributed-training command covering the common cases.	5 / 5
Workflow Clarity	Three numbered workflows plus explicit approval gates, status checks, and troubleshooting feedback loops; minor gaps where individual workflows lack inline validate-before-proceed checkpoints.	4 / 5
Progressive Disclosure	Strong section structure with a Quick Reference table and external resource links makes navigation easy, though the 520-line body is monolithic with no bundle-file references that could offload detail.	4 / 5
	Total	16 / 20 Passed

Description

92%Weight 40%Scale 1-5

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 specific, well-triggered description that clearly states both what the skill does and when to use it, with concrete GPU/storage keywords and low conflict risk. Only minor synonym coverage is missing.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — on-demand GPU clusters, git-style training jobs, specific GPU models (B200/H200/H100), persistent NFS storage, and batch training jobs — giving comprehensive coverage rather than vague language.	5 / 5
Completeness	Explicitly answers 'what' ('On-demand GPU clusters and training jobs with git-style interface') and 'when' via a concrete 'Use when you need...' clause with specific triggers.	5 / 5
Trigger Term Quality	Strong natural terms ('GPU clusters', 'training jobs', 'NFS storage', explicit GPU model names) that users would say, but a few common synonyms like 'distributed training' or 'fine-tuning' are absent.	4 / 5
Distinctiveness Conflict Risk	Clear niche tied to the TensorPool CLI and named GPU models with distinct triggers, making conflict with other skills unlikely.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (533 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/cloud-compute/tensorpool/SKILL.md
Commit: 3a6c3a9

Reviewed: 1 day 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.