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 renky1025/agent-skills --skill prompt-enhancergit clone --depth 1 https://github.com/renky1025/agent-skillsWrote 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/renky1025/agent-skills/prompt-enhancer)<a href="https://agentmods.dev/skills/renky1025/agent-skills/prompt-enhancer"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/prompt-enhancer/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/renky1025/agent-skills/prompt-enhancer"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/prompt-enhancer.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 | $0.00290 | $0.03086 |
| Opus 5 | $0.00145 | $0.01543 |
| Sonnet 5 | $0.00058 | $0.00617 |
| Haiku 4.5 | $0.00029 | $0.00309 |
Grade A, and why
prompt-enhancer 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 today.
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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
增强提示词(prompt-enhancer)
把"一句话草稿"变成"八段式生产级强指令",减少 AI 拿到任务就开干、做错方向、交出平庸结果的行为。
用法
增强提示词:<原始提示词>
- 直接贴入原始提示词即可;可追加说明:
受众 / 领域 / 我满意的例子 / 输出语言 - 权威方法论来源:
references/开发强指令.md(含第 2-12 章各领域完整模板,需要时读取对应章节)
核心方法论(摘自《开发强指令.md》第 0 章 / 第 15 章)
八条指导思想速查
| # | 思想 | 一句话动作 |
|---|---|---|
| 1 | 复述需求 | 动手前让它说"我理解的是……" |
| 2 | 先思考 | 让它先给 2-3 个候选方案及取舍,再给最终版 |
| 3 | 角色+目标 | 给专业角色,并说清什么最重要 |
| 4 | 不要什么 | 显式写"不要 XXX",比只说目的地更有效 |
| 5 | 自我批评 | 答完让它对照验收标准找最弱处并重写 |
| 6 | 不同方向 | 要"完全不同的思路"而非"再来 5 个版本" |
| 7 | 给例子 | 让它分析例子规律再创作,不模仿字面 |
| 8 | 充分上下文 | 好结果 = 上下文 + 限制 + 反馈 + 例子 + 目标 |
强指令最小可用骨架(增强的输出必须覆盖这些块)
你是一名[专业角色]。目标:[最终要解决的问题 + 什么最重要]。
上下文:
[项目 / 技术栈 / 现状 / 约束 / 代码 / 日志 / 架构]
我满意的例子(可选但强烈建议):
[例子] —— 先分析其语气、结构、节奏规律,再据此创作。
第一步 · 复述对齐(原则1):
开始前,先告诉我你理解了什么:目标 / 限制条件 / 打算怎么改。确认无误再继续。
第二步 · 先思考(原则2):
不要直接动手。先列出 2-3 个候选方案及各自取舍,再给最终版本。
执行要求:
1. [具体怎么做] 2. [覆盖哪些维度] ...
明确不要(原则4):
- 不要[具体反面行为]
完成后 · 自我批评(原则5):
批评自己的产出,找出最弱的部分,说明如何改进,然后重写。
输出与验收(原则8):
[必须交付什么 / 如何判断做对了]
如果信息不足,不要猜测。明确指出缺失信息,并基于现有证据完成能确定的部分。
工作流程
第 1 步 · 解析原稿
提取:意图(要 AI 完成什么)、产出物、目标受众、隐含约束、领域。留意原稿里的口语化/错字表达,按真实意图归一化(如"我向导的是"->"我想要的是"),不把错字带进增强版。
第 2 步 · 诊断缺口(对照骨架逐项检查)
逐项标记原稿缺失的块,重点查这些常见病:
- 无角色 / 无目标,或目标不可验证
- 口头结构要求没有固化为步骤(例:"要有主业务路径、业务能力、产品能力"只是描述,没变成执行步骤+自检)
- "不清楚就问 / 有问题随时问"太含糊,没定义问什么、什么时候问
- 无"明确不要",AI 易跑偏或虚构(编造仓库不存在的证据等)
- 无验收标准 / 交付清单
- 无防平庸机制(自我批评、不同方向)
第 3 步 · 选模板
按任务领域从下方"领域模板库"选取最贴近的骨架;跨领域任务合并多个模板的要素。
第 4 步 · 增强填充(核心技巧)
- 角色落地:给专业身份 + 判断准则(例:售前解决方案架构师,擅长把技术翻译成高层业务语言)
- 目标可验证:把"生成一张图"升级为带验收口径的目标(例:让不懂技术的客户高层 3 分钟看懂)
- 口头结构要求 -> 显式步骤 + 闭环自检:用户口述的层次/顺序要求逐条展开为执行步骤,并加"自检三层是否闭环"这类核对动作
- "不清楚就问" -> 分级澄清闸门:拆成"必须由我确认"vs"可用合理假设"两级;策略是"先出基于证据的骨架 + 分级提问清单,不硬出全量结果",避免白做也避免卡死
- 防虚构红线:凡是基于代码/事实/数据的任务,明确"只基于证据,不编造不存在的结论"
- 受众表达规范:目标受众决定了语言边界(高层->禁黑话禁技术图;工程师->保术语保精确)
- 验收清单化:输出部分写成"交付物 1/2/3...",每项可核对
- 保留原稿全部硬性要求:可增不可删、不可篡改原意
第 5 步 · 输出与自检
- 自检增强版是否覆盖骨架全部八块、是否消化了原稿每个要求点
- 输出:强指令全文(对话内 code block;用户要求保存时写 .md 到工作目录)+ 增强对照表 + 用法说明
- 若原稿信息严重不足(缺目标受众 / 产出物 / 领域),先问 1-2 个关键问题再增强;或给"默认假设版 + 待确认项清单"两段式,不阻塞交付
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.
- today Changed aed2d686c49f
- yesterday First seen · 139 lines · 290 tokens per session scan A 747652cfff72
prompt-enhancer is a skill published in the GitHub repository renky1025/agent-skills (11 stars, last pushed yesterday), licensed MIT. It adds 290 tokens to every session and 3,086 once invoked, about $0.0014 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 2026-09-10.
Other skills, from other repositories
llm-redteam-overview
LLM red team category — full AATMF v3 tactic coverage (T01–T15). Routing skill: read this first to identify which tactic applies, then load the matching sub-skill. Maps to MITRE ATLAS where overlap exists.
omh-llm-app-dev
This is a Hermes-native llm-app-dev workflow skill.
omh-model-setup
This is a Hermes-native model-setup workflow skill.
omh-model-optimization
This is a Hermes-native model-optimization workflow skill.
model-onboarding
Onboard a new model generation or sibling into oh-my-hermes: probe router recognition, research the official contract, write trait-to-counter calibration, place routing in both lanes, price from documented list only, gate machine config on a served route, prove with the gates, close with a benchmark pair. Use when a…
prompt-scanner
A scanner for text sent to an AI agent, looking for prompt injection and jailbreak attempts. Prompt injection is text that tries to override an agent's instructions; a jailbreak tries to bypass its safety limits.