Agent skill

Create Master

by xr843 in xr843/Master-skill

“基于佛教经典文献,生成特定高僧大德的 AI 教学角色”

— description from SKILL.md by xr843
MITAuto-check passed

Install Create Master

skills CLI
$ npx skills add xr843/Master-skill --skill create-master -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install xr843/Master-skill create-master --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
create-master
GitHub stars
447
Token cost
~1.6k tokens
SKILL.md length
406 words
Files
295 (incl. scripts, references)
Skills in repo
20
Repo updated
First seen
Licence
MIT

At a glance

  • Works in 7 steps: :检查运行环境 → :信息录入 → :数据采集 → …
  • SKILL.md covers 触发条件, 预置法师(直接调用,无需生成), 教学模式(多祖师协作) and 主流程(生成新法师), plus 6 more sections
  • Calls python3

About this skill

Create Master is a skill in xr843/Master-skill (447 stars). Its SKILL.md is about 1.6k tokens, with 294 other files in the folder (scripts, references). Licence: MIT.

Requirements

  • Pre-approved tools (allowed-tools): Read, Glob, Grep, Bash(python3 ${CLAUDE_SKILL_DIR}/tools/*), Bash(python3 "${CLAUDE_SKILL_DIR}/tools/*)

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. :检查运行环境
  2. :信息录入
  3. :数据采集
  4. :分析与生成
  5. 5:二阶段审查
  6. :预览与确认
  7. :写入文件

What it can do on your machine

Read from SKILL.md and the folder at commit 56196ef. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Glob
    • Grep
    • Bash(python3 ${CLAUDE_SKILL_DIR}/tools/*)
    • Bash(python3 "${CLAUDE_SKILL_DIR}/tools/*)

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/, which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Create Master loads about 1.6k tokens when it runs, and up to ~14k if it reads all its reference files. Until then it costs about 10 tokens; SKILL.md has 406 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~10
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~14k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from xr843/Master-skill at commit 56196ef, republished under its MIT licence (© xr843). 406 words, ~1,626 tokens.

Download SKILL.mdSave it as .claude/skills/create-master/SKILL.md (or your agent's skills folder). This skill also uses 294 other files; get the full folder from GitHub.
name
create-master
description
基于佛教经典文献,生成特定高僧大德的 AI 教学角色
allowed-tools
Read, Glob, Grep, Bash(python3 ${CLAUDE_SKILL_DIR}/tools/*), Bash(python3 "${CLAUDE_SKILL_DIR}/tools/*)
argument-hint
<法师名称>
version
1.0.0
user-invocable
true

Master-skill — 佛教法师教学角色生成器

本内容依据历史佛教文献生成,仅供参考学习。如需正式修行指导,请亲近善知识。

触发条件

  • /create-master 或 /create-master <法师名>
  • "帮我创建一个印光大师的教学角色"
  • "生成慧能大师的 AI Skill"
  • "我想和玄奘法师学习"

预置法师(直接调用,无需生成)

印度

  • /master-nagarjuna — 龙树菩萨(印度·中观|八宗共祖)

汉传

  • /master-xuanzang — 玄奘法师(法相唯识宗)
  • /master-kumarajiva — 鸠摩罗什(三论宗/中观)
  • /master-huineng — 慧能大师(禅宗六祖)
  • /master-zhiyi — 智顗大师(天台宗)
  • /master-fazang — 法藏大师(华严宗)
  • /master-yinguang — 印光大师(净土宗)
  • /master-ouyi — 蕅益大师(天台/净土·跨宗派)
  • /master-xuyun — 虚云老和尚(禅宗·五宗兼嗣)

藏传

  • /master-atisha — 阿底峡尊者(噶当派开祖 · 三士道 · 982-1054)
  • /master-tsongkhapa — 宗喀巴大师(格鲁派创始人 · 三主要道 · 1357-1419)
  • /master-milarepa — 米拉日巴尊者(噶举派 · 大手印 · 1052-1135)

南传

  • /master-buddhaghosa — 觉音尊者(上座部论师 · 《清净道论》· 5世纪)
  • /master-mahasi-sayadaw — 马哈希尊者(缅甸内观 · 标记法 · 1904-1982)
  • /master-ajahn-chah — 阿姜查(泰国森林禅林派 · 1918-1992)

教学模式(多祖师协作)

  • /compare-masters — 多位法师对同一问题的并列对比(横向 / 单轮)
  • /master-debate — 祖师就争议议题进行多轮交叉辩论(默认 4 轮,部分配对 5 轮;看分歧)
  • /master-curriculum — 按你的传统给出"根基→深入→精研→盲点"学修路径(纵向时序)

选择哪个模式?读 references/teaching-modes.md(含决策树与示例)。

主流程(生成新法师)

Step 0:检查运行环境

先运行 python3 "${CLAUDE_SKILL_DIR}/tools/check_deps.py"。生成器的每个工具启动时都要导入 requests、pyyaml、pypinyin,缺任何一个都会直接报 ModuleNotFoundError,连离线步骤也跑不了。退出码非 0 时停下,把它打印的安装方法原样告诉用户(系统 Python 拒绝 pip 时改用虚拟环境);不要擅自安装,那会改动用户的环境。

Step 1:信息录入

加载 ${CLAUDE_SKILL_DIR}/prompts/intake.md,3 问模式收集:①法师名称(FoJin KG 自动匹配) ②关注方面(教义/修行/讲解/全部) ③语言偏好(按传承默认)。

快捷入口、KG 匹配兜底、名称校验规则细节 → references/workflow-details.md §Step 1。

Step 2:数据采集

使用 ${CLAUDE_SKILL_DIR}/tools/sutra_collector.py --name "<法师名>" --tradition "<传承>" --output collected_data.json 从 FoJin 采集知识图谱实体、经典内容、传承术语。采集后运行 ${CLAUDE_SKILL_DIR}/tools/verify_sources.py --check-links collected_data.json,离线验证来源家族、ID 格式、声明归属与自动派生的 citation contract。

API 故障 / 超时 / 数据阈值 / 引用规则细节 → references/workflow-details.md §Step 2 + references/source-conventions.md。

Step 3:分析与生成

两阶段分析:教义(prompts/sutra_analyzer.md)→ 风格(prompts/voice_analyzer.md);按 FoJin KG 宗派标签自动选择风格规则。然后 prompts/teaching_builder.md 生成 teaching.md、prompts/voice_builder.md 生成 voice.md(4 层结构)。RAG 检索指引由 prompts/rag_instructions.md 嵌入。

宗派标签清单、Layer 0-3 含义、质量门控阈值 → references/workflow-details.md §Step 3。

Step 3.5:二阶段审查

生成器先从 sources[].type 在生成器内存中派生 citation contract,再把同一个 sources/contract 上下文交给教义准确性审查(doctrine_reviewer.md,按 citation_contract.minimum_claim_coverage 审核声明来源覆盖率) → 风格一致性(voice_reviewer.md,Layer 0 硬规则完整)。审查顺序不可颠倒。FAIL → 自动修复重审,最多 2 轮,仍 FAIL → 人工介入;最终写入的 meta.json 必须复用并再次校验同一 contract。

Step 4:预览与确认

展示 teaching.md / voice.md 结构化预览给用户。用户可要求修改特定教义、调整语气、补充主题、整体重新生成。

Step 5:写入文件

将审查通过的名称、传承、宗派、时代、语言、teaching_content、voice_content、sources 与同一 citation_contract 写入 generated-master.json。先运行 ${CLAUDE_SKILL_DIR}/tools/master_builder.py --review-digests generated-master.json 获取教义和风格两份输入哈希;两阶段审查各给出 PASS 后,在规格中写入 review.doctrine、review.voice,每份含 verdict: "PASS"、互不相同的 reviewer 标识和对应的 input_sha256,再运行 ${CLAUDE_SKILL_DIR}/tools/master_builder.py --spec generated-master.json --output "${CLAUDE_SKILL_DIR}/masters/"。生成器统一写入 masters/master-{slug}/,且 SKILL.md 的 name 为 master-{slug};随后运行 ${CLAUDE_SKILL_DIR}/tools/verify_sources.py --final-check "${CLAUDE_SKILL_DIR}/masters/master-{slug}/"。该终验离线检查四个必需文件、目录/name 一致性,以及 meta.json 的来源清单、家族 ID、声明归属和 contract;它不解析 teaching.md 自由文本,也不保证外部站点可达。

缺少审查记录或审查后改稿会在写入前失败;注册时重新核对落盘正文。哈希只证明记录对应当前内容,不证明审查结论正确。

终验通过后运行 ${CLAUDE_SKILL_DIR}/tools/master_builder.py --register "${CLAUDE_SKILL_DIR}/masters/master-{slug}":Claude Code 只加载 ~/.claude/skills/<名字>/SKILL.md,不扫 masters/,不注册就无法用 /master-{slug} 调用。按输出的 invoke 告知用户;restart_required 为真时提示重启;报 not replacing 时不得覆盖同名 skill,交由用户决定。细节与 OpenClaw 注册 → references/workflow-details.md §角色注册(按运行环境)。

Show full SKILL.md (165 more words)Show less

追加材料、纠正、管理命令

  • 追加材料:用户说"给{法师}追加{经文}"或"补充关于{主题}" → 加载 prompts/merger.md 增量合并;版本号自动 minor 递增;旧版本归档 .versions/。
  • 纠正模式:用户说"他不会这样说话/他应该更严厉" → 加载 prompts/correction_handler.md;以 ## Correction 块追加到 teaching.md / voice.md 末尾;patch 递增。
  • 管理命令:/list-masters(列出所有,标 [预置]/[自定义])、/master-rollback <slug> <version>(回滚,自动归档当前)、/delete-master <slug>(删除,预置不可删,需二次确认)。

冲突处理策略、版本号细节、用户确认文案 → references/workflow-details.md §追加材料、纠正、管理命令细则。

执行优先级(运行时)

  1. voice.md Layer 0 硬规则
  2. Correction 记录
  3. voice.md Layer 1-3
  4. teaching.md 教义内容
  5. FoJin RAG 实时检索
  6. LLM 自身知识

冲突时高优先级覆盖低优先级。示例与典型冲突场景 → references/workflow-details.md §执行优先级。

工具路由

任务工具
FoJin 数据查询${CLAUDE_SKILL_DIR}/tools/fojin_bridge.py
FoJin 实时检索${CLAUDE_SKILL_DIR}/tools/rag_query.py
经文采集${CLAUDE_SKILL_DIR}/tools/sutra_collector.py
角色生成${CLAUDE_SKILL_DIR}/tools/master_builder.py
文件写入${CLAUDE_SKILL_DIR}/tools/skill_writer.py
版本管理${CLAUDE_SKILL_DIR}/tools/version_manager.py
来源验证${CLAUDE_SKILL_DIR}/tools/verify_sources.py
教义审查${CLAUDE_SKILL_DIR}/prompts/doctrine_reviewer.md
风格审查${CLAUDE_SKILL_DIR}/prompts/voice_reviewer.md

KG 深度遍历 / 跨词典对比等 rag_query.py 不够用的场景 → references/fojin-api.md(REST API 完整参考)。

铁律(HARD-GATE)

  • NO DOCTRINAL CLAIM WITHOUT A DECLARED SOURCE CITATION. — 所有教义断言、修行指导与文本解释的引用必须解析到所选 persona 的 meta.json.sources[],来源类型必须列于 citation_contract.allowed_source_types;仅当 citation_contract.live_retrieval_allowed 为 true 时才可实时检索
  • NO FABRICATED SOURCES — 不得编造来源 ID / 引文 / 链接,所有引用必经 verify_sources.py 验证
  • NO FICTIONAL PERSONAS — 仅历史真实人物,不为虚构角色创建

完整理性化防御表、红旗清单、ETHICS.md 运行时摘要 → references/ethics-runtime.md。

敏感性边界(一句话)

不评宗派优劣 · 不宣神通感应 · 不涉政治议题 · 不代用户做重大决定 · 不替代真善知识。

涉及"祖师怎么看 XX 现代议题"边界场景 → references/ethics-runtime.md。

按需载入(progressive disclosure 路由)

触发场景读
用户问三大传统差异、宗派定位references/traditions.md
用户给出 T-/X-/SC-/Toh-/W- 引用,需验证或解析references/source-conventions.md
用户问"祖师怎么看 XX 现代议题"边界场景、AI 透明度、版权references/ethics-runtime.md
用户犹豫该用 compare / debate / curriculum 哪个references/teaching-modes.md
进入主流程 Step 1-5 任一步的细节、错误兜底、追加/纠正策略references/workflow-details.md
需要直接打 FoJin REST API(KG 深度遍历等)references/fojin-api.md
治理文档:完整 ETHICS、版权分级、Tier B 授权流程根目录 ETHICS.md

© xr843, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 294 other files (scripts, references) in the repository root of xr843/Master-skill.

  • SKILL.md
  • .claude-plugin/marketplace.json
  • .claude-plugin/plugin.json
  • .codex/INSTALL.md
  • .cursor-plugin/plugin.json
  • .gitattributes
  • .github/ISSUE_TEMPLATE/boundary_violation.yml
  • .github/ISSUE_TEMPLATE/bug_report.yml
  • .github/ISSUE_TEMPLATE/config.yml
  • .github/ISSUE_TEMPLATE/feature_request.yml
  • .github/ISSUE_TEMPLATE/new_master.yml
  • .github/PULL_REQUEST_TEMPLATE.md
  • .github/dependabot.yml
  • .github/promptfoo/package.json
  • .github/workflows
  • … and 280 more

Open the folder on GitHubat commit 56196ef

Compare with similar skills

Create Master next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Create Master compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create Master this skillxr843/Master-skill447—~1.6kAutomated safety check: PassMIT
Music Masterruvnet/ruflo74k—~270Automated safety check: PassMIT
Scrum Masteralirezarezvani/claude-skills28k1 repos~2.4kAutomated safety check: PassMIT
PPT Masterhugohe3/ppt-master58k1 repos~2.5kAutomated safety check: PassMIT
Git Mastercode-yeongyu/oh-my-openagent70k—~1.4kAutomated safety check: PassCustom licence
Git Mastercode-yeongyu/oh-my-openagent70k—~7.3kAutomated safety check: PassCustom licence

Similar skills

  • Music Master

    ruvnet/ruflo

    Run a mastering pass (LUFS loudness normalization + peak limiting) on an existing production

    74k GitHub stars~270 tokensUpdated today
    DatabasesAuto-check passed
  • Scrum Master

    alirezarezvani/claude-skills

    Advanced Scrum Master skill for data-driven agile team analysis and coaching.

    28k GitHub starsUsed in 1 repo~2.4k tokens
    Product & Project ManagementAuto-check passed
  • PPT Master

    hugohe3/ppt-master

    Generates editable PowerPoint decks, rebuilds slides from images, fills .pptx templates and polishes existing presentations through routed workflows.

    58k GitHub starsUsed in 1 repo~2.5k tokens
    Documents & OfficeAuto-check passed
  • Git Master

    code-yeongyu/oh-my-openagent

    Guides an agent through commits, rebases and history questions in Git, working from the repository's actual state and changing nothing unless you asked.

    70k GitHub stars~1.4k tokensUpdated today
    DevelopmentAuto-check passed
  • Git Master

    code-yeongyu/oh-my-openagent

    MUST USE for ANY git operations. An agent skill from code-yeongyu/oh-my-openagent.

    70k GitHub stars~7.3k tokensUpdated today
    DevelopmentAuto-check passed
  • Builds DRAFT counterparty agreements from one markdown template and a small JSON spec per party, with clauses picked by the party's role.

    275k GitHub stars~2.9k tokensUpdated 3 days ago
    Legal & ComplianceAuto-check passed

More from xr843/Master-skill

All 20 skills in this repo
  • Master Nagarjuna

    xr843/Master-skill

    A skill your agent uses when user asks about 中观, 空性, 缘起性空, 八不中道, 二谛, 世俗谛, 第一义谛, 戏论, 毕竟空, 不可得, 如幻, 离四句, 破自性, 难行道易行道, 龙树, or wants teaching in 龙树菩萨 Nāgārjuna's voice.

    447 GitHub starsUsed in 2 repos~1.8k tokens
    Auto-check passed
  • Master Ajahn Chah

    xr843/Master-skill

    A skill your agent uses when user asks about 南传佛教, 上座部, Theravada, 巴利经典, 正念 sati, 放下, 三法印, 四念处, 出入息念 anapanasati, 戒定慧, 毗婆舍那, 森林禅林派, 巴蓬寺, 阿姜查, 杜多行, 中道, or wants teaching in 阿姜查 Ajahn Chah's voice.

    447 GitHub starsUsed in 1 repo~2k tokens
    Auto-check passed
  • Master Atisha

    xr843/Master-skill

    A skill your agent uses when user asks about 藏传, 噶当派, Kadam, 三士道, 菩提道灯论, Bodhipathapradīpa, 阿底峡, Atiśa, 金洲大师, 七因果, 自他相换, 菩提心, 依止善知识, 暇满, 业果, 噶当六论, 仲敦巴, 热振寺, 藏地后弘期, or wants teaching in 阿底峡尊者 Atiśa's…

    447 GitHub starsUsed in 1 repo~2k tokens
    Auto-check passed
  • Master Buddhaghosa

    xr843/Master-skill

    A skill your agent uses when user asks about 南传, 上座部, Theravāda, 巴利, 清净道论, Visuddhimagga, 戒定慧, 四十种业处, kammaṭṭhāna, 十遍, kasiṇa, 七清净, 十六观智, 阿毗达摩, Abhidhamma, 觉音尊者, Buddhaghosa, 大寺派, Mahāvihāra, 缘起十二支…

    447 GitHub starsUsed in 1 repo~2k tokens
    Auto-check passed
  • Master Curriculum

    xr843/Master-skill

    A skill your agent uses when user asks for a sequenced learning path within a Buddhist tradition — 学修次第, 先学什么, 从哪入门, 下一步读什么, curriculum, 学习计划, 路径推荐.

    447 GitHub starsUsed in 1 repo~960 tokens
    Auto-check passed
  • Master Fazang

    xr843/Master-skill

    A skill your agent uses when user asks about 华严宗, 法界缘起, 四法界, 事事无碍, 十玄门, 六相圆融, 金师子章, 一即一切, 因陀罗网, 华严经, 五教判, or wants teaching in 法藏大师 Fazang's voice.

    447 GitHub starsUsed in 1 repo~1.8k tokens
    Auto-check passed

Questions about Create Master

How do I install Create Master in Claude Code?

Run `npx skills add xr843/Master-skill --skill create-master -a claude-code`. Or copy the skill folder (the xr843/Master-skill repository) into .claude/skills/create-master in your project. Claude Code loads it when a task matches its description.

How do I install Create Master in Codex?

Run `npx skills add xr843/Master-skill --skill create-master -a codex`. Or copy the skill folder (the xr843/Master-skill repository) into .agents/skills/create-master in your project. Codex loads it when a task matches its description.

Can I use Create Master in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add xr843/Master-skill --skill create-master -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-master, .gemini/skills/create-master, .github/skills/create-master and .opencode/skills/create-master in your project.

What does Create Master need to run?

Going by SKILL.md and its folder, Create Master needs the command-line tools its instructions call (python3). Its frontmatter pre-approves these tools: Read, Glob, Grep, Bash(python3 ${CLAUDE_SKILL_DIR}/tools/*), Bash(python3 "${CLAUDE_SKILL_DIR}/tools/*).

Does Create Master access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Create Master safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Create Master use?

Create Master is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create Master use?

About 1.6k tokens (SKILL.md is roughly 6.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 12k tokens, read only when the agent opens those files.

What are the alternatives to Create Master?

Skills that share tags, products or a category with Create Master: Music Master (ruvnet/ruflo, 74k stars), Scrum Master (alirezarezvani/claude-skills, 28k stars), PPT Master (hugohe3/ppt-master, 58k stars) and Git Master (code-yeongyu/oh-my-openagent, 70k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create Master?

xr843 (a GitHub user) maintains it in xr843/Master-skill, which has 447 GitHub stars. The repository holds 20 skills in this directory. The repository was last updated on October 5, 2026.

Source: xr843/Master-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.