ai-product

Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt engineering that scales, AI UX that users trust, and cost optimization that doesn't bankrupt you. Use when: keywords, file_patterns, code_patterns.

Quality

27%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./.agent/skills/ai-product/SKILL.md

Quality

Content

22%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This SKILL.md is essentially a table of contents with a persona preamble—it contains no actionable content of its own. The body lacks any concrete guidance, code examples, or workflow steps, delegating everything to sub-skill files that are not provided in the bundle. While the structure hints at progressive disclosure, the absence of any substantive overview or quick-start material makes the skill nearly useless on its own.

Suggestions

Add a concise quick-start section with at least one executable code example (e.g., a structured output call with validation) so the skill is actionable without navigating to sub-files.

Include a brief workflow or decision tree (e.g., 'When to use RAG vs. fine-tuning vs. prompt engineering') to provide workflow clarity and help Claude choose the right pattern.

Remove or drastically shorten the persona paragraph—it adds no actionable value and wastes tokens.

Add a 1-2 sentence description next to each sub-skill link so Claude can determine relevance without opening every file.

Dimension	Reasoning	Score
Conciseness	The persona paragraph is unnecessary fluff—Claude doesn't need to be told it has 'debugged hallucinations at 3am.' However, the overall file is short and the link list is lean.	2 / 3
Actionability	There is zero concrete guidance, no code, no commands, no examples, and no executable instructions in the body. Everything is deferred to sub-skill files that are not provided, leaving the skill itself entirely non-actionable.	1 / 3
Workflow Clarity	There is no workflow, no sequencing, no steps, and no validation checkpoints. The content is just a list of links with no indication of when or how to apply them.	1 / 3
Progressive Disclosure	The skill does attempt progressive disclosure by linking to sub-skills, and references are one level deep. However, there is no quick-start content, no overview of the actual patterns, and no bundle files were provided to confirm the referenced paths exist, making navigation uncertain.	2 / 3
	Total	6 / 12 Passed

Description

32%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description suffers from two major problems: it opens with opinionated marketing copy that wastes space and provides no selection value, and its 'Use when' clause contains unfilled template placeholders ('keywords, file_patterns, code_patterns') instead of actual trigger conditions. The middle section lists relevant topic areas but frames them as broad categories rather than specific actionable capabilities.

Suggestions

Replace the 'Use when: keywords, file_patterns, code_patterns' placeholder with actual trigger terms, e.g., 'Use when the user mentions LLM integration, RAG pipelines, embedding strategies, prompt templates, AI API costs, or building AI-powered features.'

Remove the opinionated marketing opening ('Every product will be AI-powered...') and replace with concrete actions like 'Designs RAG pipelines, implements LLM API integrations, structures prompt templates, optimizes token usage and API costs.'

Add file patterns and code patterns as explicit triggers, e.g., 'Use when working with files containing OpenAI/Anthropic client code, vector store configurations, or embedding pipelines.'

Dimension	Reasoning	Score
Specificity	It names a domain (AI/LLM integration) and lists some areas like 'RAG architecture, prompt engineering, AI UX, cost optimization,' but these are broad topic areas rather than concrete actions. The opening sentence is pure marketing fluff.	2 / 3
Completeness	The 'what' is partially addressed with topic areas but lacks concrete actions. The 'when' clause is completely broken — 'Use when: keywords, file_patterns, code_patterns' is clearly an unfilled template placeholder, providing zero actual trigger guidance.	1 / 3
Trigger Term Quality	Contains some relevant keywords like 'LLM', 'RAG', 'prompt engineering', 'AI UX', and 'cost optimization' that users might naturally mention. However, the 'Use when' clause literally says 'keywords, file_patterns, code_patterns' which are placeholder tokens, not actual trigger terms.	2 / 3
Distinctiveness Conflict Risk	The AI/LLM integration domain is somewhat specific, and terms like 'RAG architecture' and 'prompt engineering' help narrow the scope. However, the broad framing ('every product will be AI-powered') and broken trigger clause reduce distinctiveness.	2 / 3
	Total	7 / 12 Passed

Validation

90%

Warnings & errors only

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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: Dokhacgiakhoa/antigravity-ide
Commit: c4a059b

Reviewed: 12 days ago

Table of Contents

Discovery Implementation Validation

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.