learning-rate-scheduler
Learning Rate Scheduler - Auto-activating skill for ML Training. Triggers on: learning rate scheduler, learning rate scheduler Part of the ML Training skill category.
35
Quality
3%
Does it follow best practices?
Impact
93%
1.02xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./planned-skills/generated/07-ml-training/learning-rate-scheduler/SKILL.mdQuality
Discovery
7%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is severely underdeveloped, essentially serving as a label rather than a functional skill description. It lacks concrete actions, meaningful trigger terms, and explicit guidance on when to use it. The redundant trigger term and boilerplate structure suggest auto-generated content without human refinement.
Suggestions
Add specific concrete actions like 'Configure step decay, cosine annealing, warmup schedules, and cyclic learning rates for neural network training'
Include a 'Use when...' clause with explicit triggers: 'Use when the user mentions learning rate schedules, lr decay, warmup periods, or needs to adjust training dynamics'
Expand trigger terms to include natural variations: 'lr schedule', 'learning rate decay', 'warmup', 'cosine annealing', 'step decay', 'cyclical learning rate'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description only names the domain 'Learning Rate Scheduler' and mentions 'ML Training' but provides no concrete actions. There are no specific capabilities listed like 'configure step decay', 'implement warmup schedules', or 'adjust learning rates'. | 1 / 3 |
Completeness | The description fails to answer 'what does this do' beyond naming itself, and while it mentions triggers, there is no explicit 'Use when...' clause explaining when Claude should select this skill. The 'when' is essentially missing. | 1 / 3 |
Trigger Term Quality | The trigger terms are redundant ('learning rate scheduler' listed twice) and lack natural variations users might say such as 'lr schedule', 'decay rate', 'warmup', 'cosine annealing', or 'step decay'. | 1 / 3 |
Distinctiveness Conflict Risk | The term 'learning rate scheduler' is fairly specific to ML training and unlikely to conflict with unrelated skills, but within an ML skill set, it could overlap with general 'ML Training' or 'optimizer' skills without clearer boundaries. | 2 / 3 |
Total | 5 / 12 Passed |
Implementation
0%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is a placeholder template with no actual content. It describes what a learning rate scheduler skill should do without providing any concrete guidance, code examples, or actionable information. The entire content could be replaced with actual PyTorch/TensorFlow scheduler examples, common scheduler types (StepLR, CosineAnnealing, OneCycleLR), and tuning guidance.
Suggestions
Add concrete code examples for common schedulers: `torch.optim.lr_scheduler.StepLR`, `CosineAnnealingLR`, `OneCycleLR` with typical parameter values
Include a decision guide: when to use step decay vs cosine annealing vs warmup schedules based on training scenario
Provide a workflow for tuning: how to monitor learning rate effects, when to adjust, validation checkpoints
Remove all meta-description sections ('Purpose', 'When to Use', 'Capabilities') and replace with actual technical content
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is entirely filler text with no actual technical information. It explains what the skill does in abstract terms without providing any concrete guidance, wasting tokens on meta-descriptions Claude doesn't need. | 1 / 3 |
Actionability | No executable code, no specific commands, no concrete examples of learning rate schedulers. The content only describes what it claims to do without actually doing it - no PyTorch/TensorFlow scheduler examples, no parameter guidance. | 1 / 3 |
Workflow Clarity | No workflow is defined. Claims to provide 'step-by-step guidance' but contains zero actual steps. No validation checkpoints, no sequence of operations for implementing or tuning learning rate schedulers. | 1 / 3 |
Progressive Disclosure | No structure beyond boilerplate sections. No references to detailed documentation, no links to examples or advanced configurations. The 'Related Skills' section mentions a category but provides no navigation. | 1 / 3 |
Total | 4 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 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 | 9 / 11 Passed | |
- Commit
f17dd51
- 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.