CtrlK
BlogDocsLog inGet started
Tessl Logo

run-experiment

Deploy and run ML experiments on local, remote, Vast.ai, or Modal serverless GPU. Use when user says "run experiment", "deploy to server", "跑实验", or needs to launch training jobs.

66

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

High

Do not use without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

71%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.

The body is highly actionable with concrete commands and a well-sequenced, validated workflow for multi-environment GPU deployment. It is held back by some verbosity and a dangling reference to a non-existent shared file.

Suggestions

Create or remove the referenced ../shared-references/compute-env-contract.md so the body does not point at a missing file, and signal it as a one-level-deep reference rather than burying it mid-paragraph.

Tighten the "Environment contract" paragraph: replace jargon like "seeded kernel witness" and "agent-follows-doc pass" with concise concrete steps, and de-duplicate the cost-awareness rules between the workflow and Key Rules.

Move the detailed W&B integration snippet and the full CLAUDE.md example into separate reference files to keep SKILL.md an overview.

DimensionReasoningScore

Conciseness

Mostly efficient concrete commands, but padded by the dense jargon-laden "Environment contract" paragraph, repeated cost-awareness notes in both the workflow and Key Rules, and restated Modal guidance.

3 / 5

Actionability

Fully executable copy-paste bash commands for every environment (SSH, Vast.ai, Local, Modal) with specific flags, template variables, and a concrete CLAUDE.md example covering the common cases.

5 / 5

Workflow Clarity

Clearly sequenced Steps 1-7 with explicit validation checkpoints (Step 2 GPU pre-flight, Step 5 verify launch) and a Key Rules checklist; minor gaps around retry on GPU-busy or sync failure for the destructive auto-destroy path.

4 / 5

Progressive Disclosure

Good section structure, but it references ../shared-references/compute-env-contract.md which does not exist in any bundle, and inlines the full W&B integration and CLAUDE.md example that could live in separate reference files.

3 / 5

Total

15

/

20

Passed

Description

91%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.

The description is strong, clearly stating both capability and explicit trigger conditions with multilingual natural phrases. It is well-scoped to the ML experiment deployment niche with only minor overlap risk against sibling GPU skills.

DimensionReasoningScore

Specificity

Names concrete actions ("Deploy and run ML experiments") across four explicit environments (local, remote, Vast.ai, Modal serverless GPU), giving several specific deployment targets, though the action verbs themselves are limited to deploy/run.

4 / 5

Completeness

Explicitly answers both what (deploy and run ML experiments on local/remote/Vast.ai/Modal GPU) and when ("Use when user says...") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural trigger coverage including synonyms and a multilingual variant: "run experiment", "deploy to server", "跑实验", and "launch training jobs".

5 / 5

Distinctiveness Conflict Risk

Clear niche (ML experiment deployment) with distinct triggers, but minor overlap risk with closely related sibling skills it delegates to (/serverless-modal, /vast-gpu) and the generic "deploy to server" phrasing.

4 / 5

Total

18

/

20

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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
wanshuiyin/Auto-claude-code-research-in-sleep
Reviewed

Table of Contents

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.