skill-forge
Skill0 starsChanged 8 days agoMITView on GitHub
昊哥-技能熔炉。建造 Agent 技能(Skill)与 MCP 插件的原技能。当用户说"建个技能""做个skill""帮我做个插件/MCP""把这本书变成技能""把这个想法做成工具""改造/魔改这个skill或MCP""把这个技能改成我的"时使用。覆盖三种建造场景(书转技能、口述想法迭代、改造他人作品),产物平台无关,可适配扣子、dsh(DeepSeek Harness)、Claude Code 及通用 MCP 客户端。当用户提到 "build a skill" "create MCP server" "forge a skill" "turn this book into a skill" 时同样适用。
49 other projects use a version of this.
Adopted by
Most recently taken up by
- brebuilds/claude-skillstheir own version, first seen Nov 8, 2026
- yashaswi-khurana/Skillstheir own version, first seen Nov 8, 2026
- Arcanea-Labs/Starlight-Intelligence-Systemtheir own version, first seen Nov 8, 2026
- AthenDrakomin-hub/agent-skillstheir own version, first seen Nov 1, 2026
- ligj1706/ProSkillstheir own version, first seen Nov 1, 2026
- Design-of-Man/claude-skillstheir own version, first seen Oct 30, 2026
Used with
Projects with it also use sigma, agent-forge and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
sanyuan0704Skillsigma
Personalized 1-on-1 AI tutor using Bloom's 2-Sigma mastery learning. Guides users through any topic with Socratic questioning, adaptive pacing, and rich visual output (HTML dashboards, Excalidraw concept maps, generated images). Use when user wants to learn something, study a topic, understand a concept, requests tutoring, says 'teach me', 'I want to learn', 'explain X to me step by step', 'help me understand', or invokes /sigma. Triggers on: learn, study, teach, tutor, understand, master, explain step by step.
Also in 3 projects that use this
xyreinsurance119Skillagent-forge
Создание, улучшение и аудит скиллов и агентов OpenClaw. Три режима: создание скилла (11 шагов), создание агента (9 шагов, с памятью и автоулучшением), улучшение существующего (5 шагов). Triggers: 'создай скилл', 'новый скилл', 'создай агента', 'новый агент', 'улучши скилл', 'skill creator', 'agent creator', 'скиллмейкер'.
Also in 3 projects that use this
localSummerSkillskill-review
回顾当前对话,整理 skill 使用情况、代码质量反馈和迭代建议,生成待确认的更新清单
Also in 3 projects that use this
NVIDIASkilldeep-research
Queue deep, multi-step research and analysis tasks to the DeepAgents worker. Supports execution depth presets, request-specific rubrics, SubAgent delegation, and RubricMiddleware cross-validation.
Also in 3 projects that use this
GoogleSkillcode-review
Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.
Also in 3 projects that use this
Other versions49 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 49
- sanyuan0704/code-review-expert★ 3.9ksame name
- sanyuan0704/sanyuan-skills★ 3.9ksame name
- AgriciDaniel/skill-forge★ 166same name
- svy04/ballast★ 73same name
- danilokhury/Synabun★ 70same name
- motiful/skill-forge★ 45same name
- WilliamSaysX/skill-forge★ 43same name
- bjcoombs/ai-native-toolkit★ 32same name
- GodModeAI2025/skill-forge★ 18same name
- maroffo/claude-forge★ 17same name
39 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
Files
昊哥-技能熔炉(skill-forge)
建技能的技能。给熔炉三样东西之一:一本书、一个想法、或别人的作品——它按顶尖工艺标准产出 Skill 或 MCP 产物,且产物在任何主流 Agent 平台都能跑。
总原则(六条铁律,每次建造必守)
- 约束越严,产物越稳:不给产物留无限自由度。主题、布局、流程全部收进预设选项;AI 需要的不是自由,是明确边界。
- 流程代替运气:禁止"一句需求直接出产物"。每个产物必须走 澄清 → 大纲(中间产物落盘)→ 建造 → 验收 四段,每段有校准点。
- 防御性写作:先把"模型最常犯的错"列出来,再写显式检查步骤。产物里每个注意事项都必须回答"这条在防什么错"。
- 验收前置:建造前先定义"什么算对"。不接受"看起来对"——必须在真实环境调用成功、规范逐条通过、干净环境盲测触发。
- 本体平台无关:产物本体只含平台无关内容(SKILL.md / references / assets / MCP server 包),平台差异全部收进薄壳适配层。禁止把单一平台的私有机制写进本体。
- 一源一档:每次建造产出《建造档案.md》(需求原文、大纲、验收结果、版本号),落在产物目录,便于迭代追溯。
Step 0 · 三要素澄清(每次建造前必做,不澄清不动手)
向用户确认三个要素,一次最多问 3 个问题;信息缺口不影响开工时可做合理假设并显式说明:
| # | 要素 | 选项 | 为什么要问 |
| 1 | 建造场景 | A 书转技能 / B 口述想法 / C 改造他人作品 | 决定走哪份流程手册(references/forge-from-*.md) |
| 2 | 产物类型 | Skill / MCP / 不确定 | 不确定时按"产物判断矩阵"给建议(见下) |
| 3 | 目标平台 | 扣子 / dsh / Claude Code / 通用MCP客户端 / 全平台(默认) | 决定套哪层壳(references/platform-adapters.md) |
产物判断矩阵(用户不确定 Skill 还是 MCP 时):
| 输入物的本质 | 建议产物 | 理由 |
| 方法论、思维框架、写作/设计规范、流程纪律 | Skill | 知识方法类靠提示词+参考文件生效,无需外部调用 |
| 需要实时数据、外部系统动作、结构化 API 交互 | MCP | 执行类能力必须落到工具调用 |
| 大量专业知识但希望模型按需检索省 token | MCP(知识型) | 参考归藏 Generative-UI-MCP:Tool 按需加载 + Resource 常驻最小约束 |
| 混合型(方法+数据) | Skill 为主 + MCP 为辅 | 分别封装,Skill 里声明何时调 MCP |
Step 1 · 场景路由
确认场景后,先读对应手册再动手:
| 场景 | 手册 | 核心动作 |
| A 书转技能 | references/forge-from-book.md | 读原文(禁二手总结)→ 蒸馏能力点 → 建造 |
| B 口述想法 | references/forge-from-idea.md | 7问澄清访谈 → 大纲确认 → 建造 → 迭代 |
| C 改造他人作品 | references/forge-from-existing.md | 许可证审查 → 安全审计 → 解剖 → 改写 |
红线:场景 C 未完成许可证审查和安全审计前,禁止进入改写环节。
Step 2 · 建造
按产物类型读建造标准:
- Skill 产物:
references/skill-standard.md(骨架模板在assets/skill-skeleton/) - MCP 产物:
references/mcp-standard.md(骨架模板在assets/mcp-skeleton/)
Step 3 · 验收(不过闸不交付)
读 references/acceptance-checklist.md,按 P0/P1/P2/P3 分级逐项过闸。P0 全绿是最低交付线。关键验收项:
- 真实环境调用成功(目标平台实际跑一次)
- 规范逐条核对(Skill 对 Anthropic 规范 / MCP 对官方协议)
- 干净环境盲测:仅凭 description,模型能否正确触发
Step 4 · 平台适配(套壳)
读 references/platform-adapters.md。三个必配壳:
- 扣子:skill-creator 上传发布(Skill)/ plugin-creator 封装(MCP)
- dsh:skills 目录直放 / dsh 插件打包(参考 dsh-guizang-design 范式)
- Claude Code 及通用 MCP 客户端:~/.claude/skills 标准格式 / npx stdio 配置 JSON
环境适配说明
- 扣子 Agent 环境:澄清用普通对话逐条确认;产物上传走 skill-creator / plugin-creator 技能流程
- dsh 环境:澄清用普通对话;壳适配参考 cordis.patch.yml 机制
- Claude Code:澄清可用 AskUserQuestion 机制
- 不要假设某平台的专有交互机制在其他平台可用;不确定时退回普通对话确认
版本与迭代
- 产物语义化版本:v0.1.0 起步,每次实质修改进位
- 产物 SKILL.md 头部保留来源识别段(熔炉出品 + 版本 + 日期),不写入产物生成内容
- 迭代时先更新《建造档案.md》再改本体
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

