recommendation-engine
Build recommendation systems with collaborative filtering, matrix factorization, hybrid approaches. Use for product recommendations, personalization, or encountering cold start, sparsity, quality evaluation issues.
73
67%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/recommendation-engine/skills/recommendation-engine/SKILL.mdQuality
Discovery
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.
This is a strong skill description that clearly defines its scope around recommendation systems, lists specific techniques and approaches, and provides explicit trigger conditions. It uses third person voice appropriately and includes both high-level user terms and domain-specific technical terms that would help Claude correctly select this skill.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and approaches: 'collaborative filtering, matrix factorization, hybrid approaches' and specific use cases like 'product recommendations, personalization' along with specific technical challenges like 'cold start, sparsity, quality evaluation issues'. | 3 / 3 |
Completeness | Clearly answers both 'what' (build recommendation systems with collaborative filtering, matrix factorization, hybrid approaches) and 'when' (Use for product recommendations, personalization, or encountering cold start, sparsity, quality evaluation issues). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'recommendation systems', 'collaborative filtering', 'matrix factorization', 'product recommendations', 'personalization', 'cold start', 'sparsity'. These cover both high-level user terms and specific technical terms someone working in this domain would use. | 3 / 3 |
Distinctiveness Conflict Risk | Recommendation systems is a clear, well-defined niche. The specific techniques (collaborative filtering, matrix factorization) and domain-specific challenges (cold start, sparsity) make it highly distinguishable from other ML or data science skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill covers recommendation systems comprehensively but suffers from significant verbosity and redundancy—concepts like cold start, matrix factorization, and item exclusion appear multiple times. Code examples are illustrative but not fully executable due to undefined imports and missing state assignments. The structure would benefit from moving the extensive Known Issues section to a reference file and tightening the main skill to a concise overview with working code.
Suggestions
Remove redundant coverage: cold start appears in 3 places (table, dedicated section, issue #3), matrix factorization in 2 places, and item exclusion in 2 places. Consolidate each concept to a single location.
Fix code executability: store user_item_matrix in CollaborativeFilter.fit(), define ContentBasedFilter or note it as a placeholder, and replace imports from nonexistent modules with inline implementations or clearly mark them as project-specific.
Move the 7 Known Issues subsections to a reference file (e.g., references/known-issues.md) and keep only a brief bullet list in the main skill with one-line descriptions.
Add validation checkpoints to the Quick Start workflow, such as verifying matrix dimensions after step 1 and checking that recommendations are non-empty after step 3.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~200+ lines with significant redundancy. The Known Issues section repeats concepts already shown (e.g., excluding interacted items is in both CollaborativeFilter.recommend_for_user and issue #4; cold start is covered in three separate places; matrix factorization appears twice). Many issues explain problems Claude would already understand (e.g., 'Recommending items user already purchased/viewed wastes recommendation slots'). | 1 / 3 |
Actionability | Code examples are mostly concrete but several are not fully executable—CollaborativeFilter references self.user_item_matrix which is never stored in fit(), the Quick Start imports from nonexistent modules (recommendation_engine, evaluation_metrics, cold_start), and the HybridRecommender references an undefined ContentBasedFilter. The code is illustrative rather than copy-paste ready. | 2 / 3 |
Workflow Clarity | The Quick Start section provides a 5-step workflow which is helpful, but it lacks validation checkpoints (e.g., checking matrix shape, verifying model convergence, validating that recommendations are non-empty). The fallback chain in issue #3 is a good pattern, but there's no overall workflow for building, validating, and deploying a recommender with explicit verification steps. | 2 / 3 |
Progressive Disclosure | The 'When to Load References' section at the end provides good one-level-deep references to detailed files, but the main body contains extensive inline content (especially the Known Issues section with 7 detailed subsections) that should arguably be in reference files. The skill tries to be both an overview and a deep dive, undermining the progressive disclosure pattern. | 2 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- secondsky/claude-skills
- Commit
88da5ff
- 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.