prompt-engineering

A methodology for writing, structuring, testing, and improving instructions for large language models. It covers goals, context, examples, output formats, edge cases, and safety limits.

In plain words
What is it for?
Use it to design or optimize prompts, add examples and context, choose an output format, and define fallback rules or guardrails.
Why use it?
It helps turn vague instructions into prompts that are easier for a model to follow and evaluate. It also provides ways to handle unusual inputs and reduce unwanted responses.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/kxlsys/openskill/prompt-engineering
Any agent
npx skills add KxlSys/OpenSkill --skill prompt-engineering
Clone the repo
git clone --depth 1 https://github.com/KxlSys/OpenSkill

Made for: Claude Code, Codex.

Per session 21 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 419 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00021 $0.00419
Opus 5 $0.00010 $0.00210
Sonnet 5 $0.00004 $0.00084
Haiku 4.5 $0.00002 $0.00042

Measured yesterday against content hash 7a959ffe702b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prompt-engineering scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ai/prompt-engineering/SKILL.md · 41 lines

What it actually says

Prompt Engineering Methodology

Cette compétence définit un cadre pour concevoir, tester et optimiser les prompts destinés aux modèles de langage de grande taille (LLMs).

Étapes de Conception d'un Prompt

  1. Définition de l'Objectif & du Rôle :

    • Assigner un rôle ou un persona clair et précis au modèle (ex: "Vous êtes un expert en cybersécurité...").
    • Définir précisément la tâche attendue.
  2. Structuration du Prompt :

    • Fournir des instructions claires et distinctes (en utilisant du Markdown, des balises XML comme <instructions>, <context>, etc.).
    • Spécifier le format de sortie souhaité (JSON, Markdown, liste à puces, ton formel/informel).
  3. Fourniture de Contexte et d'Exemples (Few-Shot Prompting) :

    • Si possible, fournir un ou plusieurs exemples de couples entrée/sortie pour guider le modèle vers le comportement souhaité.
  4. Décomposition de la Réflexion (Chain of Thought) :

    • Inciter le modèle à détailler ses étapes de réflexion avant de donner la réponse finale (ex: "Expliquez votre raisonnement étape par étape...").
  5. Gestion des Cas Limites (Fallback & Guardrails) :

    • Ajouter des garde-fous pour éviter les hallucinations ou les comportements indésirables (ex: "Si vous ne connaissez pas la réponse, répondez simplement 'Je ne sais pas'").

Rapport de Conception de Prompt

Prompt Initial

Analyse des Résultats & Limitations

Prompt Optimisé (Version Finale)

Résultats Comparatifs

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 41 lines · 21 tokens per session scan A 7a959ffe702b

Subscribe to this mod's changes

prompt-engineering is a skill published in the GitHub repository KxlSys/OpenSkill (2 stars, last pushed 10d ago), licensed MIT. It adds 21 tokens to every session and 419 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

html-ppt

Not reveal.js / Slidev / Spectacle. A hand-crafted single HTML file — zero dependencies, zero network, double-click to open.

tsaol/awesome-claude · 0 tokens

workshop-review

Audit an AWS Workshop Studio workshop against content, structure, bilingual, asset, link, and navigation best practices. Generates a structured review report with severity levels and a pass/fail verdict. Use before publishing or after major content changes.

tsaol/awesome-claude · 50 tokens

aws-doc-review

Fact-check AWS technical documents against official AWS documentation. Extracts atomic claims, verifies each against docs.aws.amazon.com, reports findings with severity and citations. Use when the user says "review document", "fact-check", "verify this doc", "prüfe", "check for errors". Verification only — never edits…

tsaol/awesome-claude · 73 tokens

file-reference-skill

Example skill demonstrating secure file reference resolution with supporting files.

maxvaega/skillkit · 15 tokens

card-diagram

Generate card-based architecture diagrams as standalone HTML files. Warm glassmorphism style with CSS-only arrows, rounded cards, and backdrop-filter effects. Use when the user asks to create a pipeline, flow, or architecture diagram in card style.

tsaol/awesome-claude · 51 tokens

git-review

Review code changes against requirements and original design. Validates implementation correctness, design alignment, code quality, and runs tests. Use before committing or creating a PR.

tsaol/awesome-claude · 35 tokens