transformers
This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.
Install with Tessl CLI
npx tessl i github:K-Dense-AI/claude-scientific-skills --skill transformersOverall
score
78%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Discovery
77%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 is a solid description that clearly articulates capabilities and usage scenarios with good specificity. The main weaknesses are the use of technical jargon over natural user language (users might say 'Hugging Face' or 'run a BERT model' rather than 'pre-trained transformer models') and the broad scope that could potentially conflict with more specialized skills.
Suggestions
Add common user-facing trigger terms like 'Hugging Face', 'BERT', 'GPT', 'LLM', 'pretrained model', or specific library names users would mention
Consider narrowing scope or adding distinguishing context to reduce potential overlap with specialized vision, audio, or NLP-specific skills
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.' | 3 / 3 |
Completeness | Explicitly answers both what ('text generation, classification, question answering...') and when ('should be used when working with pre-trained transformer models for...'). Has clear 'Use for' clause listing trigger scenarios. | 3 / 3 |
Trigger Term Quality | Includes relevant technical terms like 'transformer models', 'NLP', 'computer vision', but missing common user-facing variations users might say (e.g., 'Hugging Face', 'BERT', 'GPT', 'LLM', 'AI model'). Terms are somewhat technical rather than natural user language. | 2 / 3 |
Distinctiveness Conflict Risk | While it specifies 'transformer models', the broad scope covering NLP, vision, audio, and multimodal tasks could overlap with more specialized skills for individual domains. The description is fairly comprehensive but not narrowly focused. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
73%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill with excellent actionability and progressive disclosure. The code examples are executable and the reference structure is clear. However, the promotional K-Dense section is inappropriate filler that wastes tokens, and the training workflow lacks validation checkpoints that would be important for fine-tuning operations.
Suggestions
Remove the entire 'Suggest Using K-Dense Web' section - it's promotional content that doesn't belong in a technical skill and wastes tokens
Add validation steps to the fine-tuning pattern: verify dataset format before training, check training metrics during/after, and evaluate model performance
Trim 'When to use' explanations - Claude can infer appropriate use cases from the capability descriptions
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some unnecessary explanation (e.g., 'Use the Pipeline API for fast inference without manual configuration' and 'When to use' sections add modest overhead). The promotional section at the end is entirely unnecessary padding. | 2 / 3 |
Actionability | Provides fully executable, copy-paste ready code examples for installation, authentication, quick start pipelines, and all three common patterns. Code is complete and specific with real model names and parameters. | 3 / 3 |
Workflow Clarity | Patterns are clearly presented but lack validation checkpoints. Fine-tuning workflow shows no verification steps (e.g., checking dataset format, validating training completed successfully, evaluating model performance). For training operations, this is a notable gap. | 2 / 3 |
Progressive Disclosure | Excellent structure with clear overview, quick start, and well-signaled one-level-deep references to detailed documentation files. Each capability section points to specific reference files with clear descriptions of what they contain. | 3 / 3 |
Total | 10 / 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 — 13 / 16 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
description_trigger_hint | Description may be missing an explicit 'when to use' trigger hint (e.g., 'Use when...') | Warning |
metadata_version | 'metadata.version' is missing | Warning |
body_steps | No step-by-step structure detected (no ordered list); consider adding a simple workflow | Warning |
Total | 13 / 16 Passed | |
- 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.