aris-serverless-modal

Run GPU workloads on Modal — training, fine-tuning, inference, batch processing. Zero-config serverless: no SSH, no Docker, auto scale-to-zero. Use when user says "modal run", "modal training", "modal inference", "deploy to modal", "need a GPU", "run on modal", "serverless GPU", or needs remote GPU compute.

Quality

82%

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

65%

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

The body is highly actionable with complete executable patterns and a clear sequenced workflow, but it is hurt by inline time-sensitive pricing, a monolithic single-file structure with no reference split, and absent error-recovery loops. Splitting detail into reference files and adding a recovery loop would lift the weaker dimensions.

Suggestions

Move the time-sensitive per-second pricing table and benchmark comparison into a separate reference file (e.g. references/pricing.md) and keep SKILL.md an overview, so stale inline figures do not bloat the context.

Add an explicit error-recovery feedback loop to the workflow (e.g. on OOM/timeout check `modal app logs`, adjust GPU type or `timeout`, then re-run) to support batch and training operations.

Split the six launcher code patterns (A–F) into a references/launchers.md file and have SKILL.md point to them, reducing the monolithic wall of inline code.

Dimension	Reasoning	Score
Conciseness	The body is largely domain-specific and useful, but it inlines a time-sensitive per-second pricing table for nine GPUs (not placed in a deprecated/old-patterns section) and repeats the card-binding security warning in both Authentication and the closing note, so it could be tightened.	2 / 3
Actionability	It provides fully executable, copy-paste-ready code across patterns A–F (e.g. a complete `modal.App`/`modal.Image` launcher) and concrete CLI commands like `modal run launcher.py` and `modal volume get`, matching the top anchor.	3 / 3
Workflow Clarity	The six-step Workflow is clearly sequenced with a required pre-run cost-estimation gate and a verify/monitor step, but it lacks an explicit error-recovery feedback loop (validate → fix → retry) for batch/training operations, which the rubric flags.	2 / 3
Progressive Disclosure	The skill is a single monolithic ~319-line file with no reference files at all; content that could be split out (six launcher patterns, the full pricing table) is inline, so it is well-organized but not progressively disclosed.	2 / 3
	Total	9 / 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.

The description is excellent: it states concrete capabilities, gives an explicit "Use when" trigger list of natural terms, answers both what and when, and occupies a clearly distinguishable Modal niche. No changes needed.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "training, fine-tuning, inference, batch processing" — matching the anchor for listing several specific concrete actions rather than naming only some.	3 / 3
Completeness	It explicitly answers both "what" (run GPU workloads on Modal, zero-config serverless) and "when" via a clear "Use when user says ..." trigger clause, the anchor for the top level.	3 / 3
Trigger Term Quality	It supplies natural phrases users would actually say — "modal run", "modal training", "deploy to modal", "need a GPU", "serverless GPU" — giving good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	The skill is anchored to a clear Modal-specific niche with distinct triggers; the generic "need a GPU" phrase is a minor risk but outweighed by the Modal-scoped framing, so it does not drop to 2.	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-serverless-modal/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.