model-checkpoint-manager
Model Checkpoint Manager - Auto-activating skill for ML Training. Triggers on: model checkpoint manager, model checkpoint manager Part of the ML Training skill category.
36
Quality
3%
Does it follow best practices?
Impact
96%
1.00xAverage 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/model-checkpoint-manager/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 essentially a placeholder that provides almost no useful information for skill selection. It only states the skill name and category without describing any capabilities, actions, or usage scenarios. The redundant trigger terms and lack of concrete details make it nearly impossible for Claude to appropriately select this skill.
Suggestions
Add specific actions the skill performs, e.g., 'Saves, loads, and manages model checkpoints during training. Supports resuming interrupted training, comparing checkpoint performance, and pruning old checkpoints.'
Include a 'Use when...' clause with natural trigger terms: 'Use when the user mentions saving model weights, resuming training, loading checkpoints, .ckpt files, .pt files, or model snapshots.'
Remove the redundant duplicate trigger term and expand with variations users would naturally say like 'save my model', 'resume from checkpoint', 'load pretrained weights'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description only names the skill ('Model Checkpoint Manager') without describing any concrete actions. There are no verbs indicating what the skill actually does - no mention of saving, loading, managing, comparing, or any other specific checkpoint operations. | 1 / 3 |
Completeness | The description fails to answer 'what does this do' beyond the name itself, and provides no 'when should Claude use it' guidance. The 'Triggers on' section just repeats the skill name rather than providing meaningful trigger scenarios. | 1 / 3 |
Trigger Term Quality | The trigger terms listed are just 'model checkpoint manager' repeated twice, which is redundant and unlikely to match natural user language. Missing common variations like 'save checkpoint', 'load weights', 'resume training', 'model snapshots', '.ckpt', '.pt files'. | 1 / 3 |
Distinctiveness Conflict Risk | The term 'Model Checkpoint Manager' is somewhat specific to ML workflows and wouldn't conflict with general document or code skills. However, without concrete actions described, it could overlap with other ML-related skills dealing with model files or training workflows. | 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 an empty template with no actual content. It contains only boilerplate descriptions of what a skill should do without any concrete guidance on model checkpoint management - no code examples, no specific techniques (saving/loading checkpoints, checkpoint frequency, storage strategies), no validation steps, and no actionable instructions.
Suggestions
Add executable code examples for saving and loading model checkpoints in PyTorch and/or TensorFlow (e.g., `torch.save(model.state_dict(), 'checkpoint.pt')`)
Include a concrete workflow for checkpoint management: when to save, naming conventions, storage location, and how to resume training from a checkpoint
Add specific guidance on checkpoint strategies: best checkpoint selection, checkpoint pruning, distributed training checkpoints
Remove all meta-description content ('This skill provides...', 'Capabilities include...') and replace with actual technical instructions
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is entirely filler with no actual technical substance. It explains what the skill does in abstract terms without providing any concrete information Claude doesn't already know. Every section is padded boilerplate. | 1 / 3 |
Actionability | No concrete code, commands, or executable guidance whatsoever. The content only describes capabilities in vague terms ('provides step-by-step guidance', 'generates production-ready code') without actually providing any of these things. | 1 / 3 |
Workflow Clarity | No workflow, steps, or process is defined. The skill claims to provide 'step-by-step guidance' but contains zero actual steps for implementing model checkpoint management. | 1 / 3 |
Progressive Disclosure | No structure beyond empty section headers. No references to detailed materials, no examples, no links to related documentation. The content is a monolithic block of placeholder text with no useful organization. | 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.