Keqian Method

by staruhub

Keqian Method is an AI Engineering skill for Claude Code, published by staruhub in ClaudeSkills.

700 stars130 forkson staruhub/ClaudeSkillsAdded 2026/08/23+1% in starsRepository updated 2026/08/13

View source on GitHub Download repo ZIP

agent-skillsai-agentsclaudeclaude-codeclaude-skillsdeep-research

Install in seconds

Install Keqian Method

Copy Keqian Method into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/staruhub/ClaudeSkills/tree/main/skills/Geek-skills-keqian-method ~/.claude/skills/Geek-skills-keqian-method

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/staruhub/ClaudeSkills.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/Geek-skills-keqian-method/SKILL.md in staruhub/ClaudeSkills
Installs to: ~/.claude/skills/Geek-skills-keqian-method
Collection: One of 20 skills cataloged from this repository
Category: AI Engineering — 2451 skills

What Keqian Method does

Keqian Method outlines an AI-native product development workflow for agent-based software work, document-driven development, quality gates, and token cost control. Use it when designing harnesses, evals, and reliable Claude Code or Cursor workflows.

Keqian Method is cataloged under AI Engineering on DirSkills. Keqian Method comes from a repository tagged agent-skills, ai-agents, claude, claude-code and claude-skills.

Documentation

README

Excerpt

克谦方法论：AI-Native产品开发实战体系

核心理念：产品人思维 × 极致单Agent × 文档驱动 × 质量门禁闭环

来源：胥克谦——从音乐教师到产品经理到AI-Native连续创业者，皮影客创始人，十几万行自建skill和脚本的harness工程实践者。

第一原则：Iron Law（铁律）

概率乘是第一性原理。

每个环节的成功率相乘决定最终质量。即使每次0.99，n=51后也不及格。因此：不追求一次完美，追求每个环节可验证、可修复、可迭代。

推论：

勤不能补拙——模型能力是底线，harness和skill只是加速器和放大器
拆到足够简单，单项任务才能收敛
每个action必须对应一个eval

第二原则：单Agent极致论

不盲目使用multi-agent。单agent做到极致，再考虑编排。

何时用单Agent（默认选择）

有先后依赖关系的任务
需要上下文连贯性的长程任务
质量要求高、不容错的核心流程

何时用并行SubAgent（例外情况）

任务间明确无依赖关系（如多角度审计出报告）
并行结果合并时不易出问题
你有能力精确控制每个subagent的上下文注入

并行的陷阱

SubAgent上下文注入是个坑：注入什么、注入多少，都需要精确控制
主Agent可能假装自己是SubAgent（实际遇到过）
并行任务中一个环节出问题，整个长任务可能报废
合并结果时容易引入不一致

实践建议： 如果不确定，选顺序执行。慢但可靠。

第三原则：文档驱动开发（SDD）

7成精力投入文档质量和harness，3成精力写代码。

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Keqian Method

What else does staruhub publish alongside Keqian Method?
Keqian Method is one of 20 skills that DirSkills catalogs from staruhub/ClaudeSkills, the repository it ships in. Its siblings there include A-Share Analyst, AI Sales Champion and C Drive Cleaner. Each one is a separate skill with its own page in this directory, installs the same way Keqian Method does, and is maintained by staruhub in that same repository. The rest of the collection is listed on the staruhub/ClaudeSkills page.
How does Keqian Method compare to other AI Engineering skills?
Keqian Method ranks #1726 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Keqian Method against them. Open each page to compare what they document and how they install.