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 yaojingang/yao-geo-skills --skill yao-geo-content-refinergit clone --depth 1 https://github.com/yaojingang/yao-geo-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/yaojingang/yao-geo-skills/yao-geo-content-refiner)<a href="https://agentmods.dev/skills/yaojingang/yao-geo-skills/yao-geo-content-refiner"><img src="https://agentmods.dev/badge/skills/yaojingang/yao-geo-skills/yao-geo-content-refiner/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/yaojingang/yao-geo-skills/yao-geo-content-refiner"><img src="https://agentmods.dev/badge/skills/yaojingang/yao-geo-skills/yao-geo-content-refiner.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 | $0.00117 | $0.01135 |
| Opus 5 | $0.00059 | $0.00567 |
| Sonnet 5 | $0.00023 | $0.00227 |
| Haiku 4.5 | $0.00012 | $0.00113 |
Grade A, and why
yao-geo-content-refiner 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 13d 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.
What it actually says
Yao GEO Content Refiner
使用场景
- 将已有文章、产品页、公众号稿、白皮书章节或 SEO 页面改造成 AI 更容易理解、抽取、引用和推荐的内容。
- 需要提升结构化、可信度、FAQ 覆盖、原子事实密度、语义信息密度和跨平台可引用性。
- 面向 DeepSeek、豆包、千问、Kimi、腾讯元宝和微信生态输出中文简体交付物。
必要输入
- 原始文章全文、目标品牌、目标问题或关键词、目标平台和目标读者。
- 品牌知识库、官网资料、产品资料、客户案例、报告、资质、已授权引用来源。
- 是否保留原文风格、是否允许新增 FAQ、是否允许新增表格、是否允许新增来源。
- 是否允许联网核验、是否有内部知识库/API/表格/CRM/BI 权限、哪些数据禁止访问或禁止写入报告。
不适用场景
- 从零写新文章、只做标题优化、全站 GEO 诊断、页面技术审计或后端归因方案。
- 没有原文和来源，却要求补充具体数据、价格、客户案例、资质或效果承诺。
- 未授权访问登录后台、付费数据库、私有 API、客户隐私数据或内部系统。
必读资料
references/content-refinement-method.mdreferences/data-access-and-verification.mdreferences/evidence-and-fact-rules.mdreferences/platform-adaptation.mdreferences/research-backed-framework.mdreferences/artifact-layout.mdreferences/quality-gates.md
执行流程
- 界定改造边界，确认可用来源和禁止扩写事实。
- 先做分析完整性检查：输入边界、来源分层、主问题、追问、评分、事实、证据、结构、语义、平台、发布和自 Review。
- 做真实数据获取与核验计划：区分公开网页、用户提供资料、授权连接器/API、内部系统和无法访问项，记录权限、时间、新鲜度和失败处理。
- 按 8 个 GEO 维度做原文评分：语义密度、结构规范性、可引用性、权威信号、可读性、鲁棒性、新颖性、跨域贡献。
- 提取原子事实卡：主体、属性、数值、时间、来源、适用边界、核验状态、可引用句。
- 补结构：直接回答、摘要、H2/H3、表格、FAQ、有序步骤和来源列表。
- 补证据、证据强度、语义实体、平台适配矩阵、发布追踪建议，删除注水、不可核验强断言和过度营销句。
- 输出四格式报告，并按
references/artifact-layout.md与references/quality-gates.md自检。
版式硬规则
- Word/PDF 不允许把宽表强行塞入页面。超过 4 列的表格必须在 Word/PDF 中转成纵向事实卡。
- Word 表格必须使用固定页面宽度和显式列宽，不依赖自动列宽。
- Word 表格总宽必须小于页面可用宽度；生成后必须检查
tblGrid宽度。 - PDF 必须渲染成 PNG 检查右侧留白，不得贴边或裁切。
- HTML 可以保留宽表，但必须有横向滚动和长文本换行。
- HTML 必须带顶部 sticky 菜单栏，页面下拉时固定跟随，菜单锚点覆盖主要报告模块。
输出契约
- GEO 改造版文章。
- 分析完整性总览。
- 真实数据获取与核验计划。
- 原文 GEO 评分表。
- 改造前后差异报告。
- 原子事实卡。
- FAQ 与同义问法。
- 语义与实体地图。
- 平台适配矩阵。
- 理论依据与改造映射。
- 证据强度与缺口。
- 页面发布版 HTML 建议。
- 发布与追踪建议。
- Word、PDF、HTML、Markdown 四格式报告。
What ships with it
27 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.
- agents/interface.yaml 766 B
- evals/expected_artifacts.json 1.4 KB
- evals/quality_cases.json 1.8 KB
- evals/rubric.md 1.6 KB
- evals/trigger_cases.json 290 B
- examples/hubspot-cn-demo/hubspot-cn-geo-content-refiner-report.docx 57 KB
- examples/hubspot-cn-demo/hubspot-cn-geo-content-refiner-report.html 40 KB
- examples/hubspot-cn-demo/hubspot-cn-geo-content-refiner-report.md 30 KB
- examples/hubspot-cn-demo/hubspot-cn-geo-content-refiner-report.pdf 48 KB
- examples/hubspot-cn-demo/quality-report.json 4.5 KB
- examples/hubspot-cn-demo/report_input.json 30 KB
- manifest.json 278 B
- references/artifact-layout.md 2.8 KB
- references/content-refinement-method.md 3.1 KB
- references/data-access-and-verification.md 2.2 KB
- references/evidence-and-fact-rules.md 2.6 KB
- references/platform-adaptation.md 1.7 KB
- references/quality-gates.md 2.0 KB
- references/research-backed-framework.md 3.4 KB
- reports/artifact-design-profile-2026-05-21.md 2.1 KB
- reports/capability-iteration-2026-05-21.md 1.5 KB
- reports/layout-overflow-review.md 1.6 KB
- reports/overall-review-2026-05-19.md 2.2 KB
- reports/overall-review-2026-05-21.md 2.2 KB
- reports/reference-scan-2026-05-21.md 2.5 KB
- scripts/render_content_refiner_bundle.py 29 KB runs code
- templates/brief-template.md 983 B
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.
- 13d ago First seen · 81 lines · 117 tokens per session scan A cf2457acbaf4
yao-geo-content-refiner is a skill published in the GitHub repository yaojingang/yao-geo-skills (742 stars, last pushed 1mo ago), licensed MIT. It adds 117 tokens to every session and 1,135 once invoked, about $0.0006 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
pipefy-process-intelligence
Use this skill when the user wants to analyze an existing pipe for improvement opportunities — automation gaps, manual bottlenecks, missing AI agents, field conditions, or adjacent processes. Acts as a process analyst: investigates, diagnoses, and improves the pipe in progressive rounds — each round delivers visible…
c
OpenPaw coordinator — routes requests to skills, manages memory, knows what's installed. Use /c for any task.
-21risk-automation
Automate 21risk tasks via Rube MCP (Composio). Always search tools first for current schemas.
-2chat-automation
Automate 2chat tasks via Rube MCP (Composio). Always search tools first for current schemas.
ably-automation
Automate Ably tasks via Rube MCP (Composio). Always search tools first for current schemas.
abstract-automation
Automate Abstract tasks via Rube MCP (Composio). Always search tools first for current schemas.