MCP Server Builder
anthropics/skills
Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.
二次元角色技能蒸馏器。输入角色名+作品名,本地检索核心资料→交叉验证→行为蒸馏→生成可运行、可追溯、可更新的角色Skill。
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install qian-gugugaga/Character_Skill_Producer csp --agent claude-codeProject 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/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .claude/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install qian-gugugaga/Character_Skill_Producer csp --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .agents/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install qian-gugugaga/Character_Skill_Producer csp --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .cursor/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install qian-gugugaga/Character_Skill_Producer csp --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .gemini/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install qian-gugugaga/Character_Skill_Producer cspInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .github/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install qian-gugugaga/Character_Skill_Producer csp --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "csp" agent skill from https://github.com/qian-gugugaga/Character_Skill_Producer/tree/main into .opencode/skills/csp/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "csp", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
csp二次元角色技能蒸馏器。输入角色名+作品名,本地检索核心资料→交叉验证→行为蒸馏→生成可运行、可追溯、可更新的角色Skill。
Csp is an agent skill from qian-gugugaga/Character_Skill_Producer. 二次元角色技能蒸馏器。输入角色名+作品名,本地检索核心资料→交叉验证→行为蒸馏→生成可运行、可追溯、可更新的角色Skill。 触发词:「生成XX的skill」「蒸馏XX」「做一个XX角色」「把XX变成skill」「造一个XX」「/csp」。 模糊需求也触发:「想聊一个傲娇角色」「有没有病娇推荐」「帮我做一个XX作品里的角色」。 依赖 Python;核心站点优先走本地脚本,网页搜索 skill / MCP / WebFetch 只作为补强和失败兜底。
Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 126 other files, including scripts and reference files (for example `CLAUDE.md`, `README.md` and `README_EN.md`).
It works with Model Context Protocol and Python. The repository describes itself as: Character Skill Producer — distill anime characters into executable agent skills. The licence is MIT.
10 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 8e17a7a. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
pythonFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Csp loads about 2.2k tokens when it runs, and up to ~7.1k if it reads all its reference files. Until then it costs about 58 tokens; SKILL.md has 555 words of instructions outside code blocks.
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.
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.
The full file from qian-gugugaga/Character_Skill_Producer at commit 8e17a7a, republished under its MIT licence (© qian-gugugaga). 555 words, ~2,249 tokens.
.claude/skills/csp/SKILL.md (or your agent's skills folder). This skill also uses 121 other files; get the full folder from GitHub.把二次元角色变成可运行的 agent 行为包。不是角色卡,不是设定集,是可执行的行为程序。
CSP 做的不是复制角色台词,而是蒸馏角色的行为操作系统。
一个好的角色 Skill 应该回答:
关键区分:捕捉的是 HOW she behaves,不是 WHAT she said。标签是给人看的,行为规则是给 AI 执行的。
CSP 的目标不是把 Wiki 重新写一遍,也不是把角色台词整理成 prompt。Wiki 告诉用户「发生过什么」,CSP 要告诉 agent「在一个新情境里,这个角色会如何活着」。
一个高质量角色 Skill 至少包含六个可运行层:
| 层 | 问题 | 失败时的样子 |
|---|---|---|
| 行为镜片 | 她先注意什么、忽略什么? | 只会复述设定 |
| 反应规则 | 什么情境下靠近、逃开、攻击、沉默? | 所有问题都同一种语气 |
| 表达 DNA | 句长、停顿、敬语、自称、情绪泄露如何组合? | 只贴口癖 |
| 关系算法 | 她如何判断善意、背叛、亲近、利用? | 对所有用户都一样热情 |
| 决策底线 | 价值冲突时先保什么、牺牲什么? | 角色被用户轻易说服 |
| 诚实边界 | 哪些不知道、哪些过期、哪些只是推测? | 硬编新剧情 |
关键原则:写得进去的是行为程序,写不进去的才保留为边界。 角色的神秘感不是靠含糊制造,而是靠承认资料、视角和推断能力的限制。
CSP 面向二次元创作者提供一套可共享的角色行为基础设施,让角色 Skill 成为聊天、同人创作和互动系统可以共同调用的行为层。
当前最直接的使用场景:
未来可以扩展到:
CSP 想建设的是一条从“资料”到“行为”再到“互动叙事”的路径。今天它可以陪用户聊天、帮助创作同人;未来它可以成为 AI 互动小说游戏和角色驱动创作工具链的一部分。
CSP 默认只采用 Highest Fidelity Mode。
暂不提供快速、省 token、轻量模式。生成时应尽可能完整检索公开资料、交叉验证关键结论、记录来源和资料日期,并把局限写进 Skill。宁可消耗更多 token 得到高保真角色,也不要为了节省上下文生成薄弱角色卡。
未来如果实现轻量模式,必须作为显式选项,不能改变默认质量标准。
CSP 依赖 Python。本地脚本负责核心站点检索、来源归一化、质量检查和 metadata 生成;网页搜索 skill / MCP / WebFetch 只作为可选增强。
检索优先级:
不得把搜索 skill 当作第一步。除非本地脚本没有覆盖该站点,或本地脚本失败并记录原因。
| 优先级 | 来源 | 示例 |
|---|---|---|
| 最高 | 用户提供的官方材料 | 设定集、访谈原文、BD 特典、官方字幕、截图 |
| 高 | 官方网站、官方角色介绍、官方剧情文本 | franchise official sites, game story text |
| 高 | 萌娘百科、Wikipedia、作品 Fandom Wiki | zh.moegirl.org.cn, wikipedia.org, fandom.com |
| 中 | Bangumi、AniDB、游戏数据库、Bilibili 高质量专栏、Anime News Network | bgm.tv, anidb.net, Bestdori, BWIKI |
| 低 | 粉丝讨论、社区解读 | 必须标注为推测 |
| 排除 | 知乎、微信公众号、百度百科 | 不可作为来源 |
重要结论至少需要两个独立来源。来源冲突时保留冲突,不要强行调和。
优先使用统一入口:
python scripts/source_search.py "角色名" --work "作品名" --mode discover
python scripts/source_search.py "角色名" --work "作品名" --sources moegirl,mediawiki萌娘百科可直接使用专用脚本:
python scripts/moegirl_api.py "角色名" --intro
python scripts/moegirl_api.py "角色名" --full
python scripts/moegirl_api.py "角色名" --search
python scripts/moegirl_api.py "角色名" --wikitextWindows 本机使用 python;Linux/macOS 用户可尝试 python3。
关键网站不能因为一次失败就标注「信息不足」。必须记录:
最终仍失败时,在 references/sources.json 和对应研究文件中记录失败项。
确认角色后,先创建目录,再开始研究:
<character-slug>/
├── SKILL.md
├── manifest.json
└── references/
├── sources.json
├── distillation.md
├── quality-report.json
└── research/
├── 01-setting.md
├── 02-personality.md
├── 03-expression.md
├── 04-relationships.md
├── 05-key-scenes.md
└── 06-media-coverage.mdSkill 必须自包含。复制整个目录就能独立使用。研究文件不完整时,不得声称生成完成。
每个生成 Skill 必须记录资料搜索日期,避免作品出续作、游戏版本更新或新活动后货不对板。
manifest.json 必填字段{
"schema_version": "1.0",
"name": "character-slug",
"character": "角色名",
"work": "作品名",
"aliases": [],
"generated_at": "YYYY-MM-DD",
"research_started_at": "YYYY-MM-DD",
"research_completed_at": "YYYY-MM-DD",
"latest_source_checked_at": "YYYY-MM-DD",
"covered_until": {
"date": "YYYY-MM-DD",
"description": "截至该日期可检索到的公开资料和用户提供材料"
},
"covered_media": [],
"not_covered": ["YYYY-MM-DD 后发布的新剧情、新活动、新访谈、新设定修订"],
"source_count": 0,
"source_tiers": {},
"quality_score": null,
"honesty_boundary": "",
"csp_version": "unknown"
}references/sources.json 要求每条来源必须记录:
idsourcetitleurlsource_tierofficialitymedia_typeretrieved_atcontent_hash,能计算则计算statuswarnings失败来源也要记录,status 写为 failed。
SKILL.md 必须包含资料时间边界角色 Skill 中必须写明:
本 Skill 的资料检索完成于:YYYY-MM-DD。
行为蒸馏基于截至该日期可检索到的公开资料和用户提供材料。
如果作品在此日期后发布新剧情、新活动、新台词、访谈或设定修订,本 Skill 可能无法反映最新内容。当用户指出「最新剧情不是这样」时,角色 Skill 不得硬拗,必须回应:
我的资料更新至 YYYY-MM-DD,可能没有覆盖之后发布的内容。如果你有最新版 CSP,或可以提供新剧情 / 新资料链接,我可以帮你更新这个 Skill;这可能会消耗一些 Token。确认:
用户只说「就做 XX」时,默认全面画像并推进;同名歧义时必须询问。
先运行本地检索:
python scripts/source_search.py "<角色名>" --work "<作品名>" --mode discover目标:
references/sources.json;候选明确时自动推进;多个候选时让用户选择。
创建目标结构,写入空的:
references/sources.jsonreferences/distillation.mdreferences/research/06-media-coverage.md并记录 research_started_at。
启动 5 个并行子 agent,但它们不再盲搜。每个 agent 必须先读取 references/sources.json 和本地检索结果,只在维度缺口明确时再外部补搜。
| Agent | 重点 | 输出文件 |
|---|---|---|
| 1 设定 | 基本信息、世界观、身份、时间线、媒体覆盖 | 01-setting.md |
| 2 性格 | 行为模式、压力反应、内在矛盾、成长弧 | 02-personality.md |
| 3 表达 | 句式、词汇、自称、敬语、口癖、台词语境 | 03-expression.md |
| 4 关系 | 重要关系、社会认知、跨作品互动 | 04-relationships.md |
| 5 名场面 | 至少 8 个关键场景、决策逻辑、压力行为 | 05-key-scenes.md |
每个研究文件必须写:
对跨媒体角色必须写 06-media-coverage.md:
游戏活动、卡面故事、区域对话可作为日常行为和关系补充,但不能覆盖主线设定,除非官方明确更新。
运行:
python scripts/merge_research.py <skill_directory>检查:
展示摘要给用户确认。调研质量决定最终 Skill 上限,缺口明显时补充研究后再蒸馏。
读取 references/distillation-framework.md,写入 references/distillation.md。
核心行为模式必须回答:
在什么情况下 → 做什么 → 为什么这样筛选标准:
目标提炼 3-7 个核心行为模式。只通过一重验证的行为降级为低置信度,不写成核心。
向用户展示:
用户确认后再组装最终 Skill。
读取 references/skill-template.md 组装 SKILL.md。
同时生成:
python scripts/generate_manifest.py <skill_directory>最终 Skill 必须包含:
运行:
python scripts/quality_check.py <skill_directory>必须检查:
manifest.json;sources.json;latest_source_checked_at;retrieved_at;SKILL.md 资料时间边界;验证结果写入 references/quality-report.json。验证结果展示给用户后才算完成。
触发语:
更新流程:
manifest.json,确认 latest_source_checked_at。sources.json 与新检索结果。references/distillation.md。SKILL.md。manifest.json 的资料日期。quality_check.py。更新时不得:
README 和交付说明可以更直观地展示生成前/生成后、角色 replay、CSP 与角色卡的差异。但不照搬“夺舍”路线。CSP 的定位是高保真行为蒸馏。
优先加入:
source_search.pysource_registry.pygenerate_manifest.pyfile_manager.py(用于备份和更新)batch_distill.py 暂缓,因为当前默认追求单个角色最高质量,不追求批量低成本生产。
可提供:
python scripts/distill_prompt.py <skill_dir>输出适合普通 ChatGPT / Claude / SillyTavern / OpenClaw 的 prompt.md。但这只是便携导出,不替代原生 Agent Skill,也不包含完整研究链。
近期只吸收「修订记录」和「用户纠错记录」,暂不做自动长期记忆。CSP 的核心是角色行为蒸馏,不是陪伴机器人运行时。
| 原则 | 一句话 |
|---|---|
| 行为 > 形容词 | 描述「做什么」而非「是什么」 |
| 证据 > 印象 | 重要结论必须可追溯 |
| 矛盾 > 一致 | 保留矛盾,这是深度的来源 |
| 语境 > 台词 | 每句经典台词必须说明场景 |
| 时间边界 > 硬拗 | 资料过期时承认边界并进入更新 |
| 口语 > 文章 | Skill 输出的是角色说话,不是论文 |
| 人味 > 完美 | 角色可以不确定、前后矛盾、沉默 |
绝不做:
生成完成后可复制整个角色目录到 .claude/skills/<name>/。修改根目录 CSP 时,必须同步 examples/csp/ 中的:
SKILL.mdreferences/scripts/避免源码版和可安装版分叉。
CSP 的多 Agent 调研、阶段检查点和质量验证借鉴了 nuwa-skill。CSP 将这种蒸馏方式适配到二次元角色行为系统:不是复刻设定,而是让角色在新情境里可信地活着。
© qian-gugugaga, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 121 other files (scripts, references) in the repository root of qian-gugugaga/Character_Skill_Producer.
Open the folder on GitHubat commit 8e17a7a
Csp 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Csp this skillqian-gugugaga/Character_Skill_Producer | 298 | — | ~2.2k | Automated safety check: Pass | MIT | |
| MCP Server Builderanthropics/skills | 180k | 63 repos | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| MCP Server BuildershareAI-lab/learn-claude-code | 78k | 4 repos | ~1.2k | Automated safety check: Pass | MIT | |
| MemPalace Setup and OperationMemPalace/mempalace | 60k | — | ~2.2k | Automated safety check: Pass | MIT | |
| LangBot Plugin Developmentlangbot-app/LangBot | 18k | — | ~3.9k | Automated safety check: Pass | Apache-2.0 | |
| Vexor CLIscarletkc/vexor | 244 | 3 repos | ~1.2k | Automated safety check: Pass | MIT |
anthropics/skills
Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.
shareAI-lab/learn-claude-code
Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.
MemPalace/mempalace
Installs and configures MemPalace as a private local palace, a shared-brain hub or a client of an existing hub, including MCP registration and version-correct initialization.
langbot-app/LangBot
Guides building, debugging and testing LangBot plugins: components, SDK calls, README and locale rules, SDK pitfalls and WebSocket-based testing.
scarletkc/vexor
Semantic file discovery via vexor. An agent skill from scarletkc/vexor.
google-antigravity/antigravity-sdk-python
Design, implement, and debug autonomous AI agents and multi-agent systems using the Google Antigravity (AGY) SDK.
Works with
二次元角色技能蒸馏器。输入角色名+作品名,本地检索核心资料→交叉验证→行为蒸馏→生成可运行、可追溯、可更新的角色Skill。. Csp is an agent skill from qian-gugugaga/Character_Skill_Producer.
Run `npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a claude-code`. Or copy the skill folder (the qian-gugugaga/Character_Skill_Producer repository) into .claude/skills/csp in your project. Claude Code loads it when a task matches its description.
Run `npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a codex`. Or copy the skill folder (the qian-gugugaga/Character_Skill_Producer repository) into .agents/skills/csp in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add qian-gugugaga/Character_Skill_Producer --skill csp -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/csp, .gemini/skills/csp, .github/skills/csp and .opencode/skills/csp in your project.
Going by SKILL.md and its folder, Csp needs the command-line tools its instructions call (python). Our summary lists: Python 3.
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.
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.
Csp 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.
About 2.2k tokens (SKILL.md is roughly 9k 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 4.8k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Csp: MCP Server Builder (anthropics/skills, 180k stars), MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars), MemPalace Setup and Operation (MemPalace/mempalace, 60k stars) and LangBot Plugin Development (langbot-app/LangBot, 18k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
qian-gugugaga (a GitHub user) maintains it in qian-gugugaga/Character_Skill_Producer, which has 298 GitHub stars. The repository was last updated on June 6, 2026.
Source: qian-gugugaga/Character_Skill_Producer on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.