prompt-engineering-patterns

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, or designing production prompt templates.

Quality

62%

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 ./skills/prompt-engineering-patterns/SKILL.md

Quality

Content

57%Weight 40%Scale 1-3

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

The skill is well-structured for progressive disclosure, with an overview body and a verified set of one-level-deep reference, asset, and script files. Its weaknesses are verbosity from restating concepts Claude already knows, non-executable example code depending on a fictional library, and the absence of an explicit validation-gated workflow.

Suggestions

Cut or move generic conceptual sections (Best Practices, Common Pitfalls, Performance Optimization, Success Metrics, Next Steps) into reference files, keeping only guidance Claude would not already infer.

Replace the fictional `prompt_optimizer` Quick Start with executable code using a real, installable library (or clearly mark it as a conceptual template), so examples are copy-paste runnable.

Add an explicit step-by-step prompt-engineering workflow with validation checkpoints (e.g. draft → run on eval set → score accuracy/consistency → refine → re-score) including a validate-fix-retry loop.

Dimension	Reasoning	Score
Conciseness	The body is organized as bullet lists rather than prose, but large sections (Best Practices, Common Pitfalls, Performance Optimization, Success Metrics, Next Steps) restate generic prompt-engineering truisms Claude already knows ("Be Specific", "Show, Don't Tell", "Remove redundant words"), so it is mostly efficient with notable removable padding rather than fully lean.	2 / 3
Actionability	There is concrete guidance (Quick Start code, Progressive Disclosure levels, integration f-strings), but the headline Quick Start imports a non-existent `prompt_optimizer` package (`PromptTemplate`, `FewShotSelector`) and the integration snippets reference undefined variables, so the code is illustrative rather than copy-paste executable.	2 / 3
Workflow Clarity	Some sequencing exists (Progressive Disclosure Levels 1-4, a vague "Clarify goals → Apply → validate outcomes" Instructions block), but there is no explicit operational workflow with validation checkpoints or a validate-fix-retry feedback loop for the core task, leaving checkpoints implicit.	2 / 3
Progressive Disclosure	The Resources section clearly signals one-level-deep references with descriptions, and all eight referenced files (five references/, two assets/, one scripts/) actually exist on disk; the SKILL.md serves as an overview pointing to detailed materials, matching the well-signaled one-level-deep anchor.	3 / 3
	Total	9 / 12 Passed

Description

67%Weight 40%Scale 1-3

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 answers both what and when with an explicit Use-when clause, but its capability phrasing is abstract and buzzword-heavy, and its trigger terms are too narrow to fully distinguish it from adjacent LLM skills. It is competent but could be tightened and broadened.

Suggestions

Replace the abstract lead "Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability" with concrete actions, e.g. "Build few-shot learning systems, apply chain-of-thought reasoning, design reusable prompt templates, and optimize prompts iteratively."

Broaden trigger terms to the natural variations users actually say, adding phrases like "few-shot learning", "chain-of-thought", "system prompts", "prompt templates", and "fixing inconsistent LLM outputs".

Narrow the broad triggers ("improving LLM outputs", "maximize LLM performance") to prompt-specific phrasing to reduce overlap with RAG or fine-tuning skills.

Dimension	Reasoning	Score
Specificity	It names the domain and some actions via the use-when clause ("optimizing prompts, improving LLM outputs, or designing production prompt templates"), but the lead "Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability" is abstract buzzword-heavy fluff rather than a list of multiple concrete actions, so it stops short of a 3.	2 / 3
Completeness	It explicitly answers both what ("Master advanced prompt engineering techniques...") and when with an explicit "Use when..." trigger clause, matching the anchor that requires both an explicit what and explicit triggers.	3 / 3
Trigger Term Quality	Natural terms a user would say appear ("optimizing prompts", "improving LLM outputs", "designing production prompt templates"), but coverage is narrow and misses common variations the body itself covers (few-shot, chain-of-thought, system prompts), so it is not a full 3.	2 / 3
Distinctiveness Conflict Risk	"Prompt engineering" is a recognizable niche, but the broad triggers "improving LLM outputs" and "maximize LLM performance" could overlap with adjacent skills (RAG, fine-tuning, general LLM-output improvement), so it is not clearly conflict-free.	2 / 3
	Total	9 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: rmyndharis/antigravity-skills
Path: skills/prompt-engineering-patterns/SKILL.md
Commit: e63f7dd

Reviewed: 13 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.