building-recommendation-systems

tessl i github:jeremylongshore/claude-code-plugins-plus-skills --skill building-recommendation-systems

Execute this skill empowers AI assistant to construct recommendation systems using collaborative filtering, content-based filtering, or hybrid approaches. it analyzes user preferences, item features, and interaction data to generate personalized recommendations... Use when appropriate context detected. Trigger with relevant phrases based on skill purpose.

27%

Overall

Validation — 81%

Implementation — 7%

Activation — 17%

SKILL.md

Review

Evals

Validation

81%

Warnings & errors only

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
metadata_version	'metadata' field is not a dictionary	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Implementation

This skill is a template-style document filled with generic boilerplate that provides no actionable guidance for building recommendation systems. It describes what it would do rather than providing executable code, concrete algorithms, or specific implementation details. The content wastes significant tokens on explanations Claude doesn't need while failing to deliver the practical value a skill should provide.

Suggestions

Replace the abstract 'Examples' section with actual executable Python code showing collaborative filtering implementation (e.g., using surprise library or sklearn's NMF)

Remove generic boilerplate sections (Prerequisites, Instructions, Output, Error Handling) that add no skill-specific value

Add concrete workflow with validation steps: data loading -> train/test split -> model training -> evaluation metrics calculation -> threshold checks

Include specific library imports and function calls rather than describing what code 'will be generated'

Dimension	Reasoning	Score
Conciseness	Extremely verbose with extensive padding explaining concepts Claude already knows (what recommendation systems are, types of filtering). Generic boilerplate sections like 'Prerequisites', 'Instructions', 'Output', and 'Error Handling' add no value and waste tokens.	1 / 3
Actionability	No executable code anywhere despite claiming to 'generate code'. Examples describe what the skill 'will do' abstractly rather than providing actual implementation. No concrete commands, libraries with usage examples, or copy-paste ready snippets.	1 / 3
Workflow Clarity	The 'How It Works' section lists vague phases without concrete steps. No validation checkpoints, no feedback loops for model evaluation, and no clear sequence for the multi-step process of building a recommendation system.	1 / 3
Progressive Disclosure	Content is organized into sections with headers, but it's a monolithic document with no references to external files for detailed implementations. The structure exists but contains mostly filler content rather than appropriately split detailed materials.	2 / 3
	Total	5 / 12 Passed

Activation

17%

This description suffers from placeholder text that provides no real guidance for skill selection. While it identifies the recommendation systems domain and mentions specific approaches, the trigger guidance is completely non-functional boilerplate. The use of 'Execute this skill empowers AI assistant' violates third-person voice conventions and adds unnecessary fluff.

Suggestions

Replace the placeholder trigger text with specific natural language phrases users would say, such as 'Use when user asks for product recommendations, movie suggestions, personalized content, or mentions building a recommender system'

Remove the fluff opening 'Execute this skill empowers AI assistant to construct' and use direct third-person voice like 'Builds recommendation systems using...'

Add concrete file types, data formats, or specific outputs (e.g., 'generates ranked item lists, similarity scores, user-item matrices') to improve specificity

Dimension	Reasoning	Score
Specificity	Names the domain (recommendation systems) and lists approaches (collaborative filtering, content-based filtering, hybrid), but the actions are somewhat vague ('analyzes', 'generate') rather than concrete specific operations.	2 / 3
Completeness	While it partially describes 'what' (building recommendation systems), the 'when' clause is completely useless boilerplate: 'Use when appropriate context detected. Trigger with relevant phrases based on skill purpose' provides zero actionable guidance.	1 / 3
Trigger Term Quality	Uses technical jargon like 'collaborative filtering' and 'content-based filtering' that users wouldn't naturally say. The phrase 'Trigger with relevant phrases based on skill purpose' is a meaningless placeholder that provides no actual trigger terms.	1 / 3
Distinctiveness Conflict Risk	The recommendation system domain is somewhat specific, but the vague trigger guidance and generic phrasing could cause overlap with other data analysis or ML-related skills.	2 / 3
	Total	6 / 12 Passed

Reviewed

17 days ago

Table of Contents

Validation Implementation Activation

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.