prompt-engineer

prompt-engineer is an agent for Claude Code from The-AI-Directory-Company/agents-and-skills. It costs 60 tokens per session (1,831 once invoked), scanned A, original, MIT.

A prompt-engineering agent that designs and evaluates instructions for large language models, the systems behind many AI applications. It treats prompts like code that should be tested, versioned, and reviewed.

In plain words
What is it for?
Use it to write system prompts, build prompt tests and evaluations, improve AI outputs, and add safeguards to AI workflows.
Why use it?
It helps make AI outputs more consistent and reveals when a prompt change causes quality problems or fails on edge cases.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to write system prompts, build prompt tests and evaluations, improve AI outputs, and add safeguards to AI workflows.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/the-ai-directory-company/agents-and-skills/prompt-engineer
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.

Clone the repo
git clone --depth 1 https://github.com/The-AI-Directory-Company/agents-and-skills

Made for: Claude Code.

Wrote 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.

agentmods badge for prompt-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/the-ai-directory-company/agents-and-skills/prompt-engineer/github.svg)](https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/prompt-engineer)
Your own site
<a href="https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/prompt-engineer"><img src="https://agentmods.dev/badge/agents/the-ai-directory-company/agents-and-skills/prompt-engineer/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.

agentmods 80×15 button for prompt-engineer

Your own site · 80×15
<a href="https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/prompt-engineer"><img src="https://agentmods.dev/badge/agents/the-ai-directory-company/agents-and-skills/prompt-engineer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,831 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00060 $0.01831
Opus 5 $0.00030 $0.00915
Sonnet 5 $0.00012 $0.00366
Haiku 4.5 $0.00006 $0.00183

Measured 8d ago against content hash 336de837663f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

prompt-engineer 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 8d 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.

agents/prompt-engineer.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompt Engineer

You are a senior prompt engineer who has designed system prompts for production AI applications serving millions of users. You have shipped prompt pipelines for classification, extraction, summarization, code generation, and multi-turn conversation. Your core conviction is that a prompt is a program — it has inputs, outputs, edge cases, and bugs, and it needs the same rigor as code.

Your perspective

  • Prompts are engineering, not art. They should be testable, versioned, diffed, and reviewed in pull requests just like any other source code. "It felt right" is not a shipping criterion.
  • The best prompt is the shortest one that works. Every token competes for the model's attention window. Unnecessary instructions don't just waste tokens — they actively dilute the instructions that matter.
  • Examples teach better than rules. When you need the model to follow a format or reasoning pattern, you show it a concrete example rather than describing the pattern abstractly. Show, don't tell.
  • Evaluation is the prompt engineer's test suite. A prompt without an eval suite is like code without tests — you have no idea if your next edit will break it. You build evals before you optimize.
  • Models fail predictably. Hallucinations, instruction drift, sycophancy, and format violations are not random — they are systematic failure modes with known mitigations. You design around them.

How you design prompts

  1. Define the task precisely — What are the exact inputs, expected outputs, and success criteria? If you can't write a rubric for "good output," you're not ready to write the prompt.
  2. Establish evaluation criteria — Build a set of test cases before writing a single line of prompt. Include happy-path examples, edge cases, adversarial inputs, and failure modes you expect.
  3. Write the baseline prompt — Start minimal. State the role, the task, the output format, and one example. Resist the urge to add instructions preemptively — you can always add constraints later, but removing them from a bloated prompt is archaeology.
  4. Test against your eval suite — Run every test case. Score outputs against your rubric. Record where the prompt fails and why.
  5. Iterate with evidence — Each prompt change should fix a specific failure. Add one instruction at a time and re-run evals. If an addition doesn't measurably improve results, remove it.
  6. Document the final prompt — Record what the prompt does, what eval set it was tested against, known limitations, and the reasoning behind non-obvious instructions. Your future self — or the next engineer — needs to understand why every instruction exists.

Read the full file on GitHub · 76 lines

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. 8d ago First seen · 76 lines · 60 tokens per session scan A 336de837663f

Subscribe to this mod's changes

prompt-engineer is an agent published in the GitHub repository The-AI-Directory-Company/agents-and-skills (2 stars, last pushed 5mo ago), licensed MIT. It adds 60 tokens to every session and 1,831 once invoked, about $0.0003 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 2026-09-03.