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 staruhub/ClaudeSkills --skill geek-skills-xuefeng-methodgit clone --depth 1 https://github.com/staruhub/ClaudeSkillsWrote 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/staruhub/claudeskills/geek-skills-xuefeng-method)<a href="https://agentmods.dev/skills/staruhub/claudeskills/geek-skills-xuefeng-method"><img src="https://agentmods.dev/badge/skills/staruhub/claudeskills/geek-skills-xuefeng-method/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/staruhub/claudeskills/geek-skills-xuefeng-method"><img src="https://agentmods.dev/badge/skills/staruhub/claudeskills/geek-skills-xuefeng-method.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 | $0.00271 | $0.03191 |
| Opus 5 | $0.00135 | $0.01596 |
| Sonnet 5 | $0.00054 | $0.00638 |
| Haiku 4.5 | $0.00027 | $0.00319 |
Grade A, and why
xuefeng-method 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.
How it starts
The opening of the file, as written. The whole thing — 283 lines — stays where its author put it; the contents beside it link to each section on GitHub.
雪峰方法论：AI-Native 产品开发实战体系
核心理念：强模型依赖 × 多专精Agent × 快速校准 × 行为审计
来源：雪峰——AI-Native连续创业者，深耕AI日历管理等AI驱动产品。 核心洞察：穷举是死循环，唯快不破才是AI-native的生存之道。
与克谦方法论（keqian-method）互为对偶： 克谦解决"如何让AI在明确边界内可靠执行"， 雪峰解决"当边界本身不确定时怎么办"。
第零步：产品类型判断（必须先做）
在选择任何开发策略之前，先判断你的产品类型。 选错方法论比没有方法论更危险。
| 类型 | 特征 | 关键判断标准 | 推荐方法 |
|---|---|---|---|
| +AI（场景依赖型） | 用户行为可枚举，AI辅助执行确定性流程 | 能列出所有合法输入输出组合 | → keqian-method |
| AI-native（强模型依赖型） | AI驱动核心决策，用户行为开放式 | 用户的下一步操作你无法预测 | → 本skill |
| 混合型 | 核心流程确定，部分环节AI-native | 能拆分出哪些模块是确定的、哪些是开放的 | → 两者结合，按模块选用 |
快速判断清单
回答以下问题，如果3个以上答"是"，你大概率是AI-native：
- 用户的输入是自由文本/语音，而非选择菜单？
- 同一输入，你希望AI给出不同风格的输出？
- 用户会因为AI的回答方式而改变自己的后续行为？
- 你无法为产品写出完整的功能测试用例集？
- 产品的核心价值在于AI的"判断"而非"执行"？
第一原则：穷举是死循环
"穷举意味着：有多少人工，就有多少智能。这是死循环。"
为什么在AI-native场景下穷举不可行
在+AI场景下，克谦说"边界内可穷举，单维度选项有限"——这是对的。 但AI-native场景的数学不一样：
用户行为空间（开放） × 模型输出空间（概率性） × 上下文状态（动态）
= 组合爆炸，不可穷举
一个日历管理能有多复杂？答案是：走AI-native路线后，非常复杂。 因为用户一旦习惯AI-native交互，就永远回不到传统模式—— 你必须持续适应用户不断演化的期望。
替代穷举的三个策略
策略1：行为模式簇（Behavioral Clusters）
不枚举每个case，而是聚类用户行为模式：
原始行为空间（不可穷举）
↓ 聚类
行为模式簇（5-15个典型模式）
↓ 每个模式簇
设计对应的AI响应策略
↓ 边界case
优雅降级到确定性逻辑
策略2：优雅降级（Graceful Degradation）
AI不确定时，回退到确定性逻辑：
AI置信度 > 阈值 → AI决策（快路径）
AI置信度 < 阈值 → 确定性回退（安全路径）
AI置信度极低 → 请求人工介入（慢路径）
策略3：概率性验收（Probabilistic Acceptance）
不用 assert output == expected，而用 check output ∈ acceptable_set：
# 传统断言式（克谦适用）
assert response == "会议安排在下午3点"
# 行为属性式（雪峰适用）
assert "下午" in response
assert contains_time(response)
assert tone_is_professional(response)
assert no_hallucinated_contacts(response)
第二原则：多养专精虾
"多养两只虾，每只都比较专业，只干一种活。出了问题找bug容易。 一个全面能干的虾，出了问题找问题非常麻烦。"
专精Agent架构
┌─ 理解Agent（NLU：解析用户意图）
│
用户输入 → 路由器 ─┼─ 执行Agent（Action：调用API/修改数据）
│
├─ 校验Agent（Verify：检查执行结果）
│
└─ 表达Agent（NLG：生成用户可见回复）
每只虾只干一种活的好处：
- 出bug时，能精确定位是哪只虾的问题
- 单独升级/替换某只虾，不影响其他
- 每只虾可以用最适合它的模型（路由策略）
What ships with it
5 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.
- 13d ago First seen · 283 lines · 271 tokens per session scan A 6e31c4e3bbbf
xuefeng-method is a skill published in the GitHub repository staruhub/ClaudeSkills (712 stars, last pushed 1mo ago), licensed MIT. It adds 271 tokens to every session and 3,191 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 2026-08-30.
Other skills, from other repositories
searchcans-rag-source-curator
Build a small, diverse, evidence-ready RAG source manifest from localized Google or Bing results, Reader extracts, direct file extraction, and optional page screenshots. Use for grounding packs, source curation, knowledge-base intake, and pre-ingestion evidence checks.
genomic-intelligence
Predict regulatory features, gene structure, and expression directly from DNA sequence using Genomic Intelligence's hosted transformer DNA language models — no local GPU or model weights. Six tasks over a REST API and a hosted MCP server (keyless public demo): promoter regions, splice donor/acceptor sites, enhancer…
bulk-rnaseq
End-to-end bulk RNA-seq orchestrator — takes raw FASTQ reads through QC and trimming (FastQC, fastp/Trim Galore), alignment and quantification (STAR, Salmon, featureCounts), assembles a gene-level counts matrix, then hands off to differential expression (pydeseq2), pathway/GSEA enrichment (pathway-enrichment), and…
umap-learn
Use UMAP-learn for nonlinear dimensionality reduction, 2D/3D embeddings, clustering preprocessing, supervised or semi-supervised UMAP, DensMAP, AlignedUMAP, and Parametric UMAP workflows.
esm
Use when working directly with the esm Python SDK, ESM3 or ESMC model IDs, Forge/Biohub inference clients, or ESMFold2 folding workflows.
glycoengineering
Analyze and engineer protein glycosylation. Scan sequences for N-glycosylation sequons (N-X-S/T), predict O-glycosylation hotspots, and access curated glycoengineering tools (NetOGlyc, GlycoShield, GlycoWorkbench). For glycoprotein engineering, therapeutic antibody optimization, and vaccine design.