weights-and-biases

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Quality

73%

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

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/ml-training/weights-and-biases/SKILL.md

Quality

Content

65%Weight 40%Scale 1-3

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 broad executable examples, but it is verbose, duplicates content already in the reference files, and lacks validation checkpoints for batch sweep operations. Trimming inline material and pointing earlier to the references would improve both conciseness and progressive disclosure.

Suggestions

Move the detailed sweeps, artifacts, and integrations sections into their reference files, keeping the body to a concise overview with early inline links.

Remove marketing/pricing content ('Users: 200,000+', GitHub stars, Pricing section) that does not aid execution.

Add a verification step for sweeps (e.g. checking sweep status or best run before/after agent runs) to support the batch workflow.

Dimension	Reasoning	Score
Conciseness	Mostly actionable but padded with marketing fluff ('200,000+ ML practitioners', '10.5k+ stars'), a Pricing section, and repeated sweep-config definitions; inline sweeps/artifacts/integrations content duplicates the reference files.	2 / 3
Actionability	Provides fully executable, copy-paste-ready code across init, logging, sweeps, artifacts, and PyTorch/Lightning/Keras/HuggingFace integrations, matching the 'fully executable' anchor.	3 / 3
Workflow Clarity	Sequences are present (init → train → log → finish; define sweep → train fn → agent) but validation checkpoints are missing, and sweeps are batch operations that lack verification steps, capping workflow clarity at 2.	2 / 3
Progressive Disclosure	Real one-level-deep references (sweeps.md, artifacts.md, integrations.md) are listed under 'See Also', but the ~600-line body is a monolith that duplicates reference content and the references are only weakly signaled at the end.	2 / 3
	Total	9 / 12 Passed

Description

82%Weight 40%Scale 1-3

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 specific and distinctive with strong natural trigger terms, but it omits an explicit 'Use when…' trigger clause, capping completeness at 2. Adding when-to-use guidance would lift it to a top score.

Suggestions

Append an explicit trigger clause, e.g. 'Use when tracking ML experiments, running hyperparameter sweeps, or managing a model registry with W&B.'

Include common user phrasings like 'experiment tracking', 'wandb', or 'training dashboards' to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Track ML experiments with automatic logging', 'visualize training in real-time', 'optimize hyperparameters with sweeps', 'manage model registry' — matching the 'multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly states what the skill does but lacks any 'Use when…' clause or equivalent explicit trigger guidance, so completeness is capped at 2 per the judging guidelines.	2 / 3
Trigger Term Quality	Covers natural terms users would say — 'ML experiments', 'training', 'hyperparameters', 'sweeps', 'model registry', 'W&B', 'MLOps' — giving good coverage of common variations.	3 / 3
Distinctiveness Conflict Risk	The W&B-specific niche ('manage model registry with W&B', 'collaborative MLOps platform') gives distinct triggers unlikely to conflict with other skills.	3 / 3
	Total	11 / 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
skill_md_line_count	SKILL.md is long (604 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/ml-training/weights-and-biases/SKILL.md
Commit: dc00e07

Reviewed: 11 days 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.