ai-product

You are an AI product engineer who has shipped LLM features to millions of users. You've debugged hallucinations at 3am, optimized prompts to reduce costs by 80%, and built safety systems that caught thousands of harmful outputs. You know that demos are easy and production is hard.

Quality

30%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/ai-product/SKILL.md

The canonical home for this skill is ai-product in sickn33/agentic-awesome-skills

Quality

Content

32%Weight 40%Scale 1-5

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

The body is well-sectioned but mostly aspirational: it lists patterns and anti-patterns as one-liners and leaves Sharp Edges solutions as empty comment stubs, providing little executable guidance and no real workflow. It reads as an outline of a skill rather than a working one.

Suggestions

Replace each pattern's one-liner and each Sharp Edges comment stub with a concrete, copy-pasteable code snippet (e.g. a Pydantic schema-validation example, a streaming consumer loop, a per-request cost-tracking helper).

Turn the material into a sequenced production workflow with explicit validation checkpoints (e.g. validate schema -> retry on parse failure -> log cost -> fallback on API error).

Remove the verbatim repeat of the description paragraph and the generic 'When to Use' boilerplate to tighten the token budget.

Dimension	Reasoning	Score
Conciseness	The opening paragraph duplicates the frontmatter description verbatim and the 'When to Use' line is generic boilerplate, giving several unnecessary padded sections that match the 'noticeably verbose' anchor.	2 / 5
Actionability	Patterns are one-line hints ('Use function calling or JSON mode with schema validation') and the Sharp Edges 'solutions' are bare comment placeholders ('# Always validate output:') with no executable code, leaving only minimal high-level guidance.	2 / 5
Workflow Clarity	There is no sequenced workflow and no validation checkpoints despite the production/safety context; the body only offers categorized lists (patterns, anti-patterns, sharp edges), matching the 'rough structure but gaps and absent validation' anchor.	2 / 5
Progressive Disclosure	The body is under 50 lines with no bundle files and is organized into clearly headed sections (Patterns, Anti-Patterns, Sharp Edges, When to Use), so the simple-skill exception applies with only minor organization gaps from the boilerplate section.	4 / 5
	Total	10 / 20 Passed

Description

27%Weight 40%Scale 1-5

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 reads as a persona vignette rather than a capability-and-trigger statement: it uses second-person voice, describes backstory instead of actions, and omits any 'Use when...' guidance. It is distinct from the bad examples only in that it names the LLM domain, but it fails to tell Claude what the skill does or when to invoke it.

Suggestions

Rewrite in third person as concrete capabilities, e.g. 'Hardens LLM features for production: validates structured output, streams responses, versions and regression-tests prompts, and adds safety/defense layers.'

Add an explicit trigger clause: 'Use when building, debugging, or shipping LLM-powered features, or when the user mentions hallucinations, prompt costs, safety filtering, or productionizing AI features.'

Drop the 3am/costs-by-80% persona color and the duplicated framing to avoid over-claims and reduce conflict with generic AI skills.

Dimension	Reasoning	Score
Specificity	The description names the LLM-product domain but only via persona backstory ('debugged hallucinations', 'built safety systems') rather than concrete skill capabilities; the second-person voice ('You are...', 'You've...') reduces specificity by one per the rubric guideline, dropping it from 2 to 1.	1 / 5
Completeness	It offers only a vague 'what' (a persona, not capabilities) and no 'when' guidance at all; the missing 'Use when...' clause and vague what fit the 'vague what and no when' anchor.	2 / 5
Trigger Term Quality	Relevant keywords appear (LLM, hallucinations, prompts, costs, safety systems) but common variations and synonyms are missing and there is no natural trigger phrasing a user would actually say, matching the 'some relevant keywords but missing variations' anchor.	3 / 5
Distinctiveness Conflict Risk	'AI product engineer' is very broad with no distinct triggers, creating high overlap risk with many other AI/LLM skills, matching the 'very broad; high overlap risk' anchor.	2 / 5
	Total	8 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

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

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/ai-product/SKILL.md
Commit: 900c504

Reviewed: 14 minutes 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.