Stats
Actions
Tags
From datascience
Apply research-backed prompt engineering techniques to improve LLM output quality. Offers multiple techniques with templates and references. Auto-triggered when crafting system prompts, agent instructions, or LLM prompts.
How this skill is triggered — by the user, by Claude, or both
Slash command
/datascience:prompt-engineeringThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Research-backed techniques. Each has a reference - paper, template, usage. Read reference before applying.
Research-backed techniques. Each has a reference - paper, template, usage. Read reference before applying.
| # | Technique | Best for | Reference |
|---|---|---|---|
| 1 | Psychological Prompting | Complex tasks, max effort (+45-115%) | references/psychological-prompting.md |
| 2 | Chain of Thought | Math, logic, debugging (+46%) | references/chain-of-thought.md |
| 3 | Chain of Draft | Token-limited reasoning (7.6% token cost) | references/chain-of-draft.md |
| 4 | Tree of Thought | Design decisions, architecture | references/tree-of-thought.md |
| 5 | Few-Shot | Structured output, classification | references/few-shot.md |
| 6 | Self-Refine | Code, documents, iterative quality | references/self-refine.md |
| 7 | Rephrase and Respond | Ambiguous requirements | references/rephrase-and-respond.md |
System prompts, agent definitions, orchestrator templates, claude -p prompts, evaluation criteria.
npx claudepluginhub stellarshenson/claude-code-plugins --plugin datascienceProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.