CtrlK
BlogDocsLog inGet started
Tessl Logo

aris-monitor-experiment

Monitor running experiments, check progress, collect results. Use when user says "check results", "is it done", "monitor", or wants experiment output.

76

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

92%

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

A well-structured, highly actionable monitoring workflow with executable commands, clear sequencing, and useful conditional fallbacks. Its main weakness is progressive disclosure: everything lives in one monolithic file with no references, and some detailed sub-procedures could be split out.

Suggestions

Move the detailed W&B python3 API blocks and the 'What to extract' checklist into a references/wandb-metrics.md file, keeping SKILL.md as a concise overview that links to it one level deep.

Consider extracting the Modal and vast.ai specifics into short reference snippets so the core monitor/collect/summarize flow stays scannable.

Add an explicit verification checkpoint step (e.g. 'Confirm each screen's status is finished/crashed before summarizing') to make the validation loop as concrete as the collection steps.

DimensionReasoningScore

Conciseness

The body is command-driven and lean — ssh/screen/vastai/modal/wandb snippets with minimal prose — and assumes Claude's competence rather than explaining concepts. Only a few light rationale sentences (e.g. the auto-review-loop note) pad it, so it stays at the lean/efficient anchor rather than dropping to 2.

3 / 3

Actionability

Provides fully executable, copy-paste-ready commands with placeholders throughout (ssh, screen -X hardcopy, vastai show instances, modal app logs, python3 wandb blocks), matching the anchor for specific examples and concrete guidance.

3 / 3

Workflow Clarity

A clearly sequenced Step 1–6 workflow with explicit conditional branches ('If hardcopy fails ...', 'If JSON results exist ...', 'Skip this step entirely if wandb is not set') and a verification rule ('If results look wrong, check training logs for errors before concluding'). The destructive/batch cap does not apply since this is read-only monitoring.

3 / 3

Progressive Disclosure

No bundle files exist and all content is inline in one ~120-line file with good section headings, but substantial sub-procedures (the detailed W&B python blocks, Modal section) that could live in one-level-deep reference files are kept inline, matching the 'content that should be separate is inline' anchor rather than a clean overview-plus-references structure.

2 / 3

Total

11

/

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, concise description that states concrete capabilities and gives explicit, natural trigger phrases covering both what and when. It uses third-person imperative voice and avoids fluff or over-claims.

DimensionReasoningScore

Specificity

Lists three concrete actions — 'Monitor running experiments, check progress, collect results' — matching the anchor for multiple specific concrete actions rather than vague language.

3 / 3

Completeness

Explicitly answers both what ('Monitor running experiments, check progress, collect results') and when ('Use when user says ...'), with an explicit 'Use when' trigger clause.

3 / 3

Trigger Term Quality

Provides natural phrases users would say — 'check results', 'is it done', 'monitor', 'wants experiment output' — giving good coverage of common variations.

3 / 3

Distinctiveness Conflict Risk

The experiment-monitoring niche with combined triggers ('check results', 'is it done', 'monitor' plus experiment context) is distinct and unlikely to fire for unrelated skills.

3 / 3

Total

12

/

12

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
OpenLAIR/dr-claw
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.