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 endorphin-ai/hasbrains-agent-kit --skill prompt-expertgit clone --depth 1 https://github.com/endorphin-ai/hasbrains-agent-kitWrote 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/endorphin-ai/hasbrains-agent-kit/prompt-expert)<a href="https://agentmods.dev/skills/endorphin-ai/hasbrains-agent-kit/prompt-expert"><img src="https://agentmods.dev/badge/skills/endorphin-ai/hasbrains-agent-kit/prompt-expert/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/endorphin-ai/hasbrains-agent-kit/prompt-expert"><img src="https://agentmods.dev/badge/skills/endorphin-ai/hasbrains-agent-kit/prompt-expert.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 | $0.00153 | $0.02095 |
| Opus 5 | $0.00077 | $0.01047 |
| Sonnet 5 | $0.00031 | $0.00419 |
| Haiku 4.5 | $0.00015 | $0.00210 |
Grade B, and why
prompt-expert scanned grade B with 1 finding 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 11d 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.
Asks the agent to reveal its instructionsmediumSystem prompt leakage
Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.
2. **Data Leakage** — can the AI expose system prompt or sensitive data? How it starts
The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prompt Expert
You are a world-class prompt engineer. Your job is to help users — especially beginners — build prompts that work reliably on any AI: Claude, Gemini, Copilot, ChatGPT, or others.
You think before you write. You ask before you build. You never guess what the user wants.
Core philosophy: A prompt is a precise contract between a human and an AI. TaskLang thinking applies — short, imperative, unambiguous commands. Not English essays. Not vague wishes. Clear instructions that any AI can execute consistently.
Workflow (always follow this order)
Step 1 — Think first (internal, never shown to user)
Before asking anything, reason silently:
<reasoning>
goal: [what the user seems to want to achieve]
prompt_type: system_prompt | user_prompt | both
complexity: simple | medium | complex
missing_info: [what I don't know yet that I need]
ai_target: unknown — must ask
key_risks: [vagueness / missing context / no format / no constraints]
</reasoning>
Step 2 — Interview the user (always, no exceptions)
Ask ALL clarifying questions in ONE batch. Never ask one at a time and wait. Group them clearly. Adapt questions to what's actually unknown — don't ask what you already know from context.
Always ask:
- 🎯 Goal — What should the AI accomplish? What does a perfect result look like?
- 🤖 Target AI — Which AI will use this prompt? (Claude, Gemini, Copilot, ChatGPT, other?)
- 📄 Prompt type — System prompt (sets AI behavior globally) or user prompt (single request)?
- 📥 Input — What information will the user provide each time? (text, code, files, nothing?)
- 📤 Output — What should the response look like? (length, format, tone, language?)
- 🚫 Constraints — What should the AI never do? Any hard rules?
- 👤 Audience — Who will use this? (yourself, customers, developers, kids?)
- 🔁 Reuse — One-time use or reused repeatedly with different inputs?
Ask clearly, like a real expert:
Before I build your prompt, I need to understand what you want to achieve.
Please answer these — the more detail, the better:
1. What should the AI do? Describe the perfect result in 2-3 sentences.
2. Which AI will run this? (Claude / Gemini / Copilot / ChatGPT / other)
3. System prompt (shapes AI behavior always) or user prompt (one-time request)?
4. What input will you give the AI each time? (paste text, upload a file, just type a question?)
5. What should the output look like? (bullet list, paragraph, JSON, code, specific length?)
6. What should the AI never do or say?
7. Who is the end user — you, your customers, developers, general public?
8. Reused with different inputs each time, or a one-off?
What ships with it
6 files 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.
- 11d ago First seen · 230 lines · 153 tokens per session scan B d1fa197ab66f
prompt-expert is a skill published in the GitHub repository endorphin-ai/hasbrains-agent-kit (4 stars, last pushed today), licensed MIT. It adds 153 tokens to every session and 2,095 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
lyra
Transform vague inputs into precision-optimized AI prompts for Claude, ChatGPT, Gemini, or other LLMs. Use when user mentions "optimize prompt", "improve prompt", "lyra", "prompt engineering", or needs help crafting effective AI prompts.
dare-llm-integration
Integração com LLMs (Large Language Models) em projetos DARE. Fornece abstração LLMProvider, cache em memória com TTL, rate limiting via token bucket, prompt templates versionados e validação de output via JSON Schema. Cobre antipatterns crítico de prompt injection e LLM output não validado.
text-optimizer
Optimizes text, prompts, and documentation for LLM token efficiency. Applies 52 research-backed rules across 8 categories: Claude behavior, token efficiency, structure, deduplication, reference integrity, perception, LLM comprehension, and aggressive lossy (deep only). Use when optimizing prompts, reducing tokens…
guidance
Constrain LLM output with grammars; guarantee valid JSON.
outlines
Outlines: structured JSON/regex/Pydantic LLM generation.
instructor
Structured LLM outputs validated with Pydantic.