senior-prompt-engineer

World-class prompt engineering skill for LLM optimization, prompt patterns, structured outputs, and AI product development. Expertise in Claude, GPT-4, prompt design patterns, few-shot learning, chain-of-thought, and AI evaluation. Includes RAG optimization, agent design, and LLM system architecture. Use when building AI products, optimizing LLM performance, designing agentic systems, or implementing advanced prompting techniques.

Quality

45%

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 ./bundled/skills/senior-prompt-engineer/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.

The body is a generic senior-engineer template dominated by buzzword bullet lists and concepts Claude already knows, with only scattered executable commands and no real sequenced workflows or validation checkpoints; references exist but are underused as a disclosure mechanism.

Suggestions

Cut the generic best-practice, responsibility, and tech-stack bullet lists that restate what Claude already knows; keep only skill-specific, non-obvious guidance to improve token efficiency.

Turn the deployment/training command lists into explicit sequenced workflows with validation checkpoints (e.g., validate config, deploy, verify rollout, monitor) so batch/destructive operations have feedback loops.

Move the detailed pattern and reference content into the existing reference files and have the body point to them concisely, rather than duplicating vague summaries inline.

Dimension	Reasoning	Score
Conciseness	The ~220-line body is padded with generic concepts Claude already knows ('Test-driven development', 'Code reviews and pair programming', 'Monitor everything critical', 'Mentor junior engineers', 'Stay current with research') and buzzword bullet lists, matching the 'verbose; explains concepts Claude knows; padded with unnecessary context' anchor rather than lean, value-dense content.	1 / 3
Actionability	There is some concrete executable guidance (the Quick Start and Common Commands give real invocations like 'python scripts/prompt_optimizer.py --input data/' and 'helm upgrade service ./charts/'), but the bulk of the body is vague abstract description ('Fault-tolerant design', 'Load balancing', 'Establish best practices') with reference sections summarized by generic bullets, so it is incomplete rather than fully actionable.	2 / 3
Workflow Clarity	No real multi-step processes are sequenced with validation checkpoints; deployment and training operations ('kubectl apply -f k8s/', 'helm upgrade', 'python scripts/train.py') appear as bare command lists with no validation or feedback loops, matching the 'steps unclear or missing; no validation mentioned for risky operations' anchor.	1 / 3
Progressive Disclosure	Bundle files referenced in the body (references/prompt_engineering_patterns.md, references/llm_evaluation_frameworks.md, references/agentic_system_design.md, and scripts/) are real and one level deep, but the SKILL.md itself carries a large inline wall of generic bullet content that belongs in references, fitting the 'some structure but content that should be separate is inline' anchor.	2 / 3
	Total	6 / 12 Passed

Description

67%

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 clearly states both capabilities and an explicit 'Use when' trigger, but it leans on buzzword-laden, high-level phrasing rather than crisp concrete actions, leaving it somewhat generic and at risk of overlapping with neighboring AI/ML skills.

Suggestions

Replace abstract framing like 'World-class prompt engineering skill' and 'Expertise in Claude, GPT-4' with concrete, observable actions the skill performs (e.g., 'Drafts, refactors, and evaluates prompts; builds few-shot and chain-of-thought templates; benchmarks LLM outputs').

Tighten trigger terms toward natural user phrasings (e.g., 'optimize my prompts', 'add few-shot examples', 'build a RAG pipeline') and drop jargon-only listings that inflate conflict risk.

Narrow the scope so it is clearly distinguishable from general AI-product or ML-engineering skills.

Dimension	Reasoning	Score
Specificity	The description names the domain and some actions ('optimizing LLM performance, designing agentic systems, implementing advanced prompting techniques') but these are high-level abstract verbs rather than multiple specific concrete actions, matching the 'names domain and some actions, but not comprehensive' anchor and falling short of the concrete-action anchor at 3.	2 / 3
Completeness	It explicitly answers both what the skill does and when to use it via the clause 'Use when building AI products, optimizing LLM performance, designing agentic systems, or implementing advanced prompting techniques', satisfying the explicit-trigger requirement for the top anchor.	3 / 3
Trigger Term Quality	It includes relevant terms a user might say ('building AI products', 'RAG optimization', 'agentic systems') but mixes in technical jargon and buzzword framing ('chain-of-thought', 'few-shot learning', 'world-class', 'Expertise in Claude, GPT-4'), missing common variations and triggering the buzzword penalty, so it lands at 'some relevant keywords but missing common variations' rather than full coverage.	2 / 3
Distinctiveness Conflict Risk	The prompt-engineering/LLM-optimization niche is somewhat specific, but broad triggers like 'building AI products' and 'optimizing LLM performance' could overlap with adjacent skills, matching the 'somewhat specific but could still overlap' anchor rather than a clearly conflict-free niche.	2 / 3
	Total	9 / 12 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
referenced_paths_exist	Referenced path issues: 3 missing	Warning

	Total	15 / 16 Passed

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/senior-prompt-engineer/SKILL.md
Commit: f627ab5

Reviewed: about 6 hours 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.