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.
npx skills add k-vaca/prompt-engineering-skill --skill prompt-engineeringgit clone --depth 1 https://github.com/k-vaca/prompt-engineering-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/k-vaca/prompt-engineering-skill/prompt-engineering)<a href="https://agentmods.dev/skills/k-vaca/prompt-engineering-skill/prompt-engineering"><img src="https://agentmods.dev/badge/skills/k-vaca/prompt-engineering-skill/prompt-engineering/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/k-vaca/prompt-engineering-skill/prompt-engineering"><img src="https://agentmods.dev/badge/skills/k-vaca/prompt-engineering-skill/prompt-engineering.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00215 | $0.08286 |
| Opus 5.5 | $0.00086 | $0.03314 |
| Sonnet 5.5 | $0.00043 | $0.01657 |
| Haiku 4.5 | $0.00021 | $0.00829 |
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 1mo ago.
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.
How it starts
The opening of the file, as written. The whole thing — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Professional Prompt Engineering
Three standing rules:
- Specify precisely, delimit unambiguously, keep patterns consistent. Phrasing tricks do not matter; these do.
- Match technique to task type and target model. Never apply a technique by habit.
- Treat every technique here as a hypothesis to validate on the target task. Trust an eval over this document.
Application strength. Apply firmly and do not second-guess: few-shot non-transfer, temperature-0 limits, eval-driven iteration. Apply as a strong default and override only on eval evidence: everything in Principle 4 for reasoning models. Apply as a default and drop the moment an eval disagrees: the Anatomy section and Principles 5–8. This skill has never been measured against an eval of its own; its rules are literature- and vendor-derived, so your eval outranks it. The Anatomy section and Principles 5–8 (delimiters, placement, output contracts, agentic reminders, instruction hierarchy) are vendor-reported and unverified: apply them as defaults, drop them when an eval disagrees. Sources live in references/evidence.md; load it only when the user asks for sources or effect sizes. Findings verified August 2026; re-check specifics on models released later.
Before writing: read the request, then profile the task
Assume the user is a non-expert in the domain their prompt targets, until the request shows otherwise. Read their words as intent, not as a spec to preserve. When the request uses domain terms correctly and specifies precisely, match that level and keep their structure. When it does not, supply the expertise they lack.
Build the prompt that serves the inferred goal, not the one they literally wrote. Add what they omitted (domain concepts, constraints, terminology, output contract, edge-case handling), cut what works against the goal, and restructure freely. Every change must serve the inferred intent, not your preference.
Resolve ambiguity by choosing, not interrogating. Pick the most reasonable reading, build to it, and state the interpretation you chose plus the main alternative in one line. Ask only when an ambiguity is consequential enough that guessing wrong wastes real effort. Never hand a non-expert a checklist of jargon they cannot answer.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 277 lines · 215 tokens per session scan A 6c7674bcd31a
prompt-engineering is a skill published in the GitHub repository k-vaca/prompt-engineering-skill (7 stars, last pushed 2mo ago), licensed MIT. It adds 215 tokens to every session and 8,286 once invoked, about $0.0009 per session on Opus 5.5. 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.
Other skills, from other repositories
google-adk
Best practices for building AI agents with Google's Agent Development Kit (ADK) in Python, covering agent design, tools, sessions, memory, artifacts, evaluation, and deployment. Use when building LLM agents or multi-agent systems with ADK, defining ADK tools, wiring up sessions/state/memory, working with ADK…
anthropic-claude-development
Expert guidance for Anthropic Claude API development including Messages API, tool use, prompt engineering, and building production applications with Claude models.
openai-api-development
Expert guidance for OpenAI API development including GPT models, Assistants API, function calling, embeddings, and best practices for production applications.
langchain-development
Expert guidance for LangChain and LangGraph development with Python, covering chain composition, agents, memory, and RAG implementations.
ai-product
Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production.
image-prompt
Use when the user wants to turn a short idea into a rich, production-grade image-generation prompt — posters, landing-page or UI mockups, ads, editorial layouts, photoreal scenes, game screenshots, logos, or illustrations. Builds ONE structured, copy-paste-ready prompt optimized for gpt-image-2 by default, following…