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 KimYx0207/Kim_Service --skill meta-skill-creatorgit clone --depth 1 https://github.com/KimYx0207/Kim_ServiceWrote 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/kimyx0207/kim_service/meta-skill-creator)<a href="https://agentmods.dev/skills/kimyx0207/kim_service/meta-skill-creator"><img src="https://agentmods.dev/badge/skills/kimyx0207/kim_service/meta-skill-creator/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/kimyx0207/kim_service/meta-skill-creator"><img src="https://agentmods.dev/badge/skills/kimyx0207/kim_service/meta-skill-creator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00245 | $0.03252 |
| Opus 5 | $0.00122 | $0.01626 |
| Sonnet 5 | $0.00049 | $0.00650 |
| Haiku 4.5 | $0.00024 | $0.00325 |
Grade A, and why
meta-skill-creator 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 12d 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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
元技能创建器
用这个技能把可重复工作流做成能运行、能验收、能迁移的技能包。产物不是“更长的提示词”，而是一个默认入口很轻、细节按需加载、校验方法可复现的能力包。
核心契约
- 从用户真实工作流出发，不从个人偏好的模板出发。
- 写文件前先证明这件事值得技能化；一次性任务不要硬做成技能包。
SKILL.md只保留路由面：触发、第一动作、渐进加载、硬停止和验证。- 详细研究、产品设计、包计划、触发评测、验收方法和模板放到
references/、assets/、evals/、scripts/。 - 需要补全模糊意图、让用户多次选择或先锁 MVP 再批量生成的 skill，必须设计 Codex
request_user_input/ Claude CodeAskUserQuestion等宿主原生决策面；Markdown 选择卡只能是降级等待界面，不能冒充真实确认。 - 图片、视频、演示文稿、文档、报表、仪表盘或其他渲染产物，优先检查 Image2 / 宿主原生能力；MCP、脚本、SVG、静态预览是降级或辅助路线，必须有降级证据。
- 外部材料只能抽象成质量原则；不要复制别人的命名、页面结构、视觉系统、提示词、示例或商业话术。
- 当候选 Skill 要影响公开受众的理解、偏好或行动，或要生成多页视觉内容时，必须先让 Skill 自己研究受众需求、公开讨论和同类供给，再形成“机会研究 -> 受众决策 -> 创意系统”。不要把流量研究变成用户必填问题，也不要把单一平台的标题/封面套路抽到通用层。
- 每次交付都要闭环：记录运行证据、评审发现、
writeback/proposal/none-with-reason/blocked决策，并把可复用学习写回对应 reference、template、validator、eval 或 example。
能力分层
完整包是能力库，不是所有候选 Skill 的固定模板。按任务风险和交付阶段选择强度：
core：所有候选都要明确目标、触发、输入、输出、执行步骤契约、边界和最小验证。目标必须说明谁在什么压力下得到什么最终产物以及如何验收；每个步骤必须有输入、动作、可观察输出、进入下一步的条件和失败路线；边界必须覆盖非目标、权限、事实来源、副作用、停止与降级。conditional：只有领域陌生、事实会变化、产物可视化、面向公开受众、需要多轮决策或依赖宿主工具时，才加载深度研究、内容机会、创意决策板、多模态或交互式 MVP 模块。release：只有声称可分发、可销售、可迁移或明显优于基线时，才强制执行 clean-session 基线、宿主证明、人工验收和完整闭环。
不得因为完整包里存在某个模板，就要求每个候选 Skill 填满它；也不得因为当前任务简单，就删除以后命中条件时需要的能力。
渐进加载
只读当前阶段需要的文件：
| 当前阶段 | 读取 | 何时使用 |
|---|---|---|
| 意图与领域研究 | references/intent-domain-research.md、assets/domain-research-brief-template.md |
所有候选先写最小领域研究简报；只有领域陌生、当前事实会变、用户要求 deep research、涉及公开兼容/安全/商业就绪声明时升级为深度研究。检索后证据仍不足则标 research-needed。内容营销或公开传播类技能再加载内容机会研究。 |
| 表面与产物链 | references/experience-surface-model.md |
Skill 要产出文件、媒体、截图、演示文稿、报告、仪表盘或其他可见产物。 |
| 产品化设计 | references/product-design.md |
需要决定多个产物、用户旅程、3 分钟可见结果、首次公开表面或商业交付时。简单单一工作流不强制。 |
| 运行契约 | references/skill-contract.md |
写入或改动候选技能包前使用。 |
| 来源抽象边界 | references/source-abstraction-boundary.md |
使用外部案例、内部样例或竞品材料时，先抽象再重写。 |
| 包计划 | assets/package-plan-template.md |
编辑前映射 SKILL.md、参考文件、资产、脚本、评测和示例。 |
| 多模态/工具路线 | references/multimodal-tooling.md、assets/multimodal-prompt-brief-template.md |
Skill 依赖 Image2、宿主媒体能力、MCP、本地渲染器或脚本；先记录本地能力清单和多模态简报。多页视觉产物还要写每页角色、画面、文字区、禁止遮挡区和页面连续性。 |
| 交互式决策与 MVP 门 | references/interactive-mvp-product-skill.md |
Skill 要在模糊意图下弹多决策、锁定风格/封面/首屏/MVP 后批量生产，或要跨 Codex、Claude Code 宿主复用确认流程。 |
| 评测设计 | references/evaluation-method.md、evals/trigger-eval.json |
core 至少设计触发正反例、两个真实任务和一个边界任务。 |
| 发布级运行证明 | assets/trigger-run-record-template.json、assets/host-event-anchor-template.json、assets/host-attestation-template.json、assets/reviewer-attestation-template.json |
仅在 release 层设计真实 with/without clean-session、目录外运行/人工锚点和基线对比。 |
| 发布门禁 | references/release-gate.md、assets/acceptance-run-template.md |
声称可以分发、迁移或交付给别人使用前。 |
| 闭环治理 | references/closed-loop-governance.md、assets/loop-run-record-template.md |
验收、失败、复盘、漂移、发布准备或用户要求“闭环”时；必须写明运行记录和写回决策。 |
What ships with it
39 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.
- assets/acceptance-run-template.md 3.9 KB
- assets/domain-research-brief-template.md 2.7 KB
- assets/host-attestation-template.json 985 B
- assets/host-event-anchor-template.json 753 B
- assets/loop-run-record-template.md 1.4 KB
- assets/multimodal-prompt-brief-template.md 1.5 KB
- assets/package-plan-template.md 2.0 KB
- assets/product-design-board-template.md 1.8 KB
- assets/reviewer-attestation-template.json 820 B
- assets/skill-contract-template.md 3.5 KB
- assets/trigger-run-record-template.json 1.3 KB
- CHANGELOG.md 635 B
- docs/images/alipay.jpg 163 KB
- docs/images/contact-qr.png 825 KB
- docs/images/wechat-pay.jpg 70 KB
- evals/trigger-eval.json 2.9 KB
- examples/example-input.md 290 B
- examples/example-output.md 5.2 KB
- LICENSE 1.0 KB
- NOTICE 220 B
- README.md 3.7 KB
- references/closed-loop-governance.md 5.3 KB
- references/evaluation-method.md 10 KB
- references/evidence.md 5.3 KB
- references/experience-surface-model.md 6.2 KB
- references/intent-domain-research.md 11 KB
- references/interactive-mvp-product-skill.md 6.6 KB
- references/multimodal-tooling.md 4.8 KB
- references/product-design.md 16 KB
- references/release-gate.md 4.9 KB
- references/skill-contract.md 8.9 KB
- references/source-abstraction-boundary.md 1.7 KB
- scripts/check_acceptance_runs.py 50 KB runs code
- scripts/check_closed_loop.py 3.9 KB runs code
- scripts/check_meta_skill_package.py 12 KB runs code
- scripts/prepare_acceptance_run.py 11 KB runs code
- scripts/run-acceptance.sh 866 B runs code
- scripts/run-baseline.sh 860 B runs code
- scripts/test_acceptance_runs.py 30 KB runs code
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.
- 12d ago First seen · 113 lines · 245 tokens per session scan A 6a7d23152128
meta-skill-creator is a skill published in the GitHub repository KimYx0207/Kim_Service (169 stars, last pushed 29d ago), licensed MIT. It adds 245 tokens to every session and 3,252 once invoked, about $0.0012 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…