skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
diskd-ai/prompting78 installs

prompting

Prompt engineering guidance for writing and improving LLM prompts. Use when asked to (1) write a prompt for a specific task, (2) review or improve an existing prompt, (3) design system prompts for AI assistants, (4) structure prompts for specific output formats (JSON, XML, markdown), or (5) apply prompt engineering techniques like few-shot, chain-of-thought, or role prompting.

How do I install this agent skill?

npx skills add https://github.com/diskd-ai/prompting --skill prompting
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    This skill is a purely educational resource containing prompt engineering guides and templates. It does not include any executable code, scripts, or network capabilities.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

  • Runlayerwarn

    2/5 files flagged

What does this agent skill do?

Prompt Engineering

Guide for crafting effective prompts for large language models (Claude, GPT, Gemini, Llama, etc.).

Workflow

  1. Clarify the goal: What task should the prompt accomplish?
  2. Choose techniques: Select from references/techniques.md
  3. Structure the prompt: Apply appropriate format
  4. Add constraints: Specify requirements and boundaries
  5. Test and refine: Iterate based on outputs

Writing a New Prompt

Start with this template:

[Context/Role - optional]
[Task - required]
[Constraints/Requirements - as needed]
[Output format - as needed]
[Examples - for complex tasks]

Minimal prompt (simple tasks):

Summarize this article in 3 bullet points.

Structured prompt (complex tasks):

You are a senior code reviewer.

Review this code for:
- Security vulnerabilities
- Performance issues
- Maintainability concerns

Format your response as:
## Summary
[1-2 sentences]

## Issues
- [severity]: [description]

## Recommendations
[prioritized list]

Improving an Existing Prompt

Diagnose issues:

ProblemSolution
Output too vagueAdd specific constraints or examples
Wrong formatSpecify output structure explicitly
Missing detailsUse chain-of-thought or decomposition
Inconsistent resultsAdd few-shot examples
Off-topic responsesStrengthen role/context framing

Improvement checklist:

  • Is the task clear and unambiguous?
  • Are constraints specific (not "be concise" but "under 100 words")?
  • Does output format match intended use?
  • Would examples clarify expectations?

Quick Reference: Techniques

TechniqueWhen to Use
Few-shotSpecific format/style needed
Chain-of-thoughtComplex reasoning, math, analysis
Role promptingDomain expertise, specific tone
Task decompositionMulti-step workflows
ConstraintsPrecise requirements

See references/techniques.md for detailed patterns and examples.

Quick Reference: Output Formats

FormatWhen to Use
XML tagsComplex prompts, clear section boundaries
JSONProgrammatic parsing, structured data
MarkdownHuman-readable reports, documentation

See references/structured.md for format patterns.

System Prompts

For designing AI assistant behavior, see references/system-prompts.md.

Key sections:

  • Identity and role definition
  • Behavioral guidelines
  • Constraints and boundaries
  • Output format defaults

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/diskd-ai/prompting/prompting">View prompting on skillZs</a>