Prompt Engineering

This is a solid reference for structured prompt improvement. It covers the fundamentals like few-shot learning, chain-of-thought reasoning, and template systems with actual examples you can adapt. The progressive disclosure pattern is especially practical: start simple, add constraints only when needed, then layer in examples. What makes this useful is the focus on systematic testing and iteration rather than guessing. You get clear before-and-after prompt versions showing how small changes affect output quality. The instruction hierarchy and error recovery patterns are worth bookmarking if you're building production agents that need consistent behavior across edge cases.

skills.sh installs: 1.1K
GitHub stars: 43.1K
last updated: Jul 14, 2026

View on GitHub

Install

$ npx -y skills add sickn33/antigravity-awesome-skills --skill prompt-engineering --agent claude-code

Installs into .claude/skills of the current project.

Files

Prompt Engineering Patterns

Advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability.

Core Capabilities

1. Few-Shot Learning

Teach the model by showing examples instead of explaining rules. Include 2-5 input-output pairs that demonstrate the desired behavior. Use when you need consistent formatting, specific reasoning patterns, or handling of edge cases. Additional examples may help or hurt; evaluate their relevance and token cost.

Example:

Extract key information from support tickets:

Input: "My login doesn't work and I keep getting error 403"
Output: {"issue": "authentication", "error_code": "403", "priority": "high"}

Input: "Feature request: add dark mode to settings"
Output: {"issue": "feature_request", "error_code": null, "priority": "low"}

Now process: "Can't upload files larger than 10MB, getting timeout"

2. Chain-of-Thought Prompting

Ask for a concise rationale, assumptions, calculations or independently checkable evidence suitable for the task. Do not request hidden internal reasoning or assume a fixed percentage improvement; effects depend on model, task and evaluation.

Example:

SKILL.md on GitHub