aris-compute-guard

Mandatory pre-flight compute resource check before running experiments. Detects whether local/remote GPU or compute resources are actually available. If resources are unavailable, STOPS the experiment pipeline immediately and reports to the user — preventing the model from hallucinating fake experiment results. Use when: about to run experiments, deploy training, or any GPU-intensive task.

Quality

88%

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

77%

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

A highly actionable and well-sequenced guard skill with executable commands and a clear decision gate, but it is verbose (redundant rule restatements and long templates) and monolithic with no progressive disclosure to reference files.

Suggestions

Collapse the CRITICAL RULE section into the intro and remove the duplicated STOP/REPORT wording to reduce token cost.

Trim the Step 4 and Step 5 message templates to a compact schema the model can fill, rather than full boilerplate prose.

Move the per-environment detection blocks (Vast.ai, Modal, remote SSH) into a reference file under ./scripts or ./references and link one level deep, improving progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly actionable and efficient, but the CRITICAL RULE section restates the mandatory-stop warning already in the intro, and the Step 4/Step 5 report templates plus the closing Rules section add padding that could be tightened.	2 / 3
Actionability	Provides fully executable bash commands for each environment (nvidia-smi, torch MPS/CUDA probes, ssh, vastai, modal) with explicit availability/unavailability criteria, matching the copy-paste-ready anchor.	3 / 3
Workflow Clarity	A clearly sequenced 5-step workflow with an explicit Decision Gate checkpoint and a feedback path (compute unavailable -> Step 4 stop/report, compute available -> Step 5 proceed), matching the validation-checkpoint anchor.	3 / 3
Progressive Disclosure	Content is a single monolithic ~180-line SKILL.md with no bundle files or one-level-deep references; the per-environment command blocks could be split out, so it sits at the some-structure-but-inline anchor.	2 / 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 with concrete actions, an explicit trigger clause, and a clear niche. It uses third-person voice throughout and avoids vague fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Detects whether local/remote GPU or compute resources are actually available", "STOPS the experiment pipeline immediately and reports to the user", and "preventing the model from hallucinating fake experiment results" — matching the multi-action anchor.	3 / 3
Completeness	It explicitly answers both what it does (pre-flight compute check, stop pipeline, report) and when to use it via an explicit "Use when:" trigger, satisfying the top anchor.	3 / 3
Trigger Term Quality	The "Use when: about to run experiments, deploy training, or any GPU-intensive task" clause covers natural phrases a user would say, with good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	The compute-resource-guard niche with GPU/experiment-training triggers is distinct and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/aris-compute-guard/SKILL.md
Commit: 6da7f7c

Reviewed: about 7 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.