CtrlK
BlogDocsLog inGet started
Tessl Logo

ai-engineer

Build production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations.

23

Quality

12%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/antigravity-ai-engineer/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Discovery

25%

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 like a marketing pitch rather than a functional skill selector, relying heavily on buzzwords ('production-ready', 'enterprise AI integrations', 'intelligent agents') without concrete specificity. It completely lacks a 'Use when...' clause, making it difficult for Claude to know when to select this skill over others. The extremely broad scope covering multiple AI domains creates high conflict risk with other skills.

Suggestions

Add an explicit 'Use when...' clause with specific trigger scenarios, e.g., 'Use when the user asks to build RAG pipelines, configure vector databases, create LLM-powered agents, or integrate AI APIs like OpenAI or Anthropic into applications.'

Replace vague buzzwords with concrete actions, e.g., instead of 'enterprise AI integrations' say 'connects to OpenAI, Anthropic, and Pinecone APIs; configures embedding pipelines; builds retrieval chains with LangChain or LlamaIndex.'

Narrow the scope or clearly delineate sub-domains to reduce conflict risk — a skill that covers LLM apps, RAG, agents, multimodal AI, and enterprise integrations is too broad to be distinctive.

DimensionReasoningScore

Specificity

Names the domain (LLM applications, RAG systems, agents) and lists some actions (vector search, multimodal AI, agent orchestration, enterprise AI integrations), but these are more like buzzword categories than concrete specific actions. It doesn't describe what it actually does with these things (e.g., 'configures vector databases', 'implements retrieval pipelines').

2 / 3

Completeness

Describes 'what' at a high level but completely lacks any 'when' clause or explicit trigger guidance. There is no 'Use when...' or equivalent statement telling Claude when to select this skill, which per the rubric should cap completeness at 2, and since the 'what' is also somewhat vague/buzzwordy, this falls to 1.

1 / 3

Trigger Term Quality

Includes relevant keywords like 'RAG', 'LLM', 'vector search', 'agents', and 'multimodal AI' that users might mention, but misses common variations and natural phrasing like 'chatbot', 'embeddings', 'retrieval augmented generation', 'AI pipeline', 'langchain', 'llamaindex', or specific framework names users would actually say.

2 / 3

Distinctiveness Conflict Risk

The description is extremely broad, covering LLM applications, RAG, agents, vector search, multimodal AI, and enterprise integrations — this could easily conflict with any number of more specific AI/ML skills. Terms like 'production-ready' and 'enterprise AI integrations' are generic enough to overlap with many other skills.

1 / 3

Total

6

/

12

Passed

Implementation

0%

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

This skill reads as a persona description or capability catalog rather than actionable instructions. It is overwhelmingly verbose, listing dozens of tools, frameworks, and concepts Claude already knows, while providing zero executable code, no concrete workflows, and no structured references. To be useful, it needs to be fundamentally restructured around specific, actionable tasks with real code examples and clear validation steps.

Suggestions

Replace the exhaustive capability lists with 2-3 concrete, executable code examples for the most common tasks (e.g., a production RAG pipeline setup, an agent workflow with LangGraph) including complete, runnable code.

Rewrite the 'Instructions' section as a detailed, sequenced workflow with explicit validation checkpoints and error recovery steps, especially for multi-step operations like RAG system setup or agent deployment.

Extract the detailed tool/framework listings into separate reference files (e.g., MODELS.md, VECTOR_DBS.md, AGENT_FRAMEWORKS.md) and keep SKILL.md as a concise overview with clear navigation links.

Remove sections that describe Claude's persona or knowledge ('Behavioral Traits', 'Knowledge Base', 'Purpose') and replace them with specific, actionable constraints and decision criteria (e.g., 'When to use Pinecone vs pgvector: use pgvector when...').

DimensionReasoningScore

Conciseness

Extremely verbose and padded with information Claude already knows. The bulk of the content is exhaustive lists of tools, frameworks, models, and concepts (e.g., listing every vector database, every embedding model, every agent framework) that provide no actionable guidance. The 'Capabilities' section reads like a resume, not instructions. The 'Behavioral Traits' and 'Knowledge Base' sections describe Claude's persona rather than teaching it how to do something new.

1 / 3

Actionability

There is no concrete, executable code, no specific commands, no examples with inputs/outputs, and no copy-paste-ready guidance anywhere in the skill. The 'Instructions' section is four vague bullet points ('Clarify use cases', 'Design the AI architecture'). The 'Example Interactions' are just prompt strings with no corresponding responses or implementations.

1 / 3

Workflow Clarity

The four-step 'Instructions' workflow is extremely vague ('Clarify use cases, constraints, and success metrics', 'Implement with monitoring, safety, and cost controls') with no validation checkpoints, no feedback loops, no error recovery steps, and no concrete sequencing. For a skill covering complex multi-step operations like RAG pipelines and agent orchestration, this is wholly insufficient.

1 / 3

Progressive Disclosure

The content is a monolithic wall of text with no references to external files, no bundle files, and no layered structure. Hundreds of lines of capability listings are inline rather than being split into focused reference documents. There is no navigation structure or signposting to deeper resources.

1 / 3

Total

4

/

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.

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

10

/

11

Passed

Repository
boisenoise/skills-collections
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.