persona-distill
v0.4.0Persona Distillation Ecosystem — a parameterized schema library + generation/evaluation/scheduling/debate toolkit for distilling personas (people or rule systems) into self-contained Claude Code skills. 5 skills, 9 persona schemas, 19 reusable components (incl. execution-profile via Klein-RPD/CDM 4-sweep), 9-phase pipeline with v0.4.0 security hardening (consent attestation, untrusted-corpus discipline, fingerprint verification, self-containment linter, rubric config lock, corpus access declaration).
By XRenSiuLicense: MIT2 GitHub starsUpdated 3 days ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 5 skill or MCP entries
- Source updated
- Sep 21, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install persona-distill for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install persona-distill@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/XRenSiu/claude-code-forgeClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/persona-distill/.
Plugin files
├── .claude-plugin/plugin.json├── skills/distill-collector/SKILL.md├── skills/distill-meta/SKILL.md├── skills/persona-debate/SKILL.md├── skills/persona-judge/SKILL.md└── skills/persona-router/SKILL.md
Included Skills5
多模态语料采集脚手架 + 4 个可运行解析器。把 iMessage / 邮件 / Twitter 归档 / 通用文本直接转成 Markdown 放进 persona skill 的 knowledge/ 目录; WeChat / QQ / Feishu / Slack / Dingtalk / Telegram / 音视频 / OCR 仍是规范 + 第三方工具指针。 Use when: (1) 需要为 distill-meta 准备多模态语料, (2) 想把跨平台聊天记录归档成统一 Markdown, (3) 需要把音频/视频访谈转写成文本, (4) 手动整理语料前想先看 CLI 契约和目录规范。 Triggers: "collect corpus", "采集语料", "ingest chat history", "transcribe interview", "import wechat", "import imessage", "parse mbox", "parse twitter archive"
Persona Distillation 主编排器。根据用户输入生成一个**自包含、可独立运行**的 persona skill 产物, 基于 9 种参数化 schema + 19 个可复用组件 + 9 阶段工作流(含迭代深化、冲突检测、CDM 执行画像提取、 自包含与指纹校验、dog-food 验证)。v0.4.0 引入 consent 闸门 + untrusted-corpus 分隔符 + corpus access 声明 + rubric config 区间锁等 7 项安全强化。 Use when: (1) 用户请求蒸馏一个人/领域/规则体系, (2) 从聊天记录/公开资料构建 persona skill, (3) 想要一个可被其他 Claude Code 项目复用的人格 skill。 Triggers: "distill 张三", "蒸馏乔布斯", "build a persona skill for X", "帮我做一个 xx 的 persona", "distill-meta", "persona distill"
编排 2-5 个已安装的 persona skill 围绕同一问题做结构化辩论,产出真正的多视角对话(而非单 persona 的独白)。 Use when: (1) 问题需要多视角对照, (2) 用户点名多个 persona 一起聊, (3) 单一 persona 给出的答案过于单薄。 Triggers: "persona debate", "多 persona 辩论", "debate this question with X, Y, Z", "多视角讨论"
Persona Skill 质量评估器。对任何 persona skill 执行 3 项测试 + 12 维 rubric + 密度评分, 产出可机读的 validation-report.md 作为质量闸门。 Use when: (1) distill-meta Phase 4 自动回调, (2) 用户主动评估已有 persona skill, (3) CI 中验证 persona skill 更新是否退化。 Triggers: "evaluate persona", "评估 persona skill", "persona quality check", "persona judge"
跨 persona 智能调度器。当用户装了多个 persona skill 时,router 读取它们的 manifest.json, 根据当前问题推荐最合适的 1-3 个 persona 来回答。 Use when: (1) 已装 10+ 个 persona skill, (2) 不确定该问哪个 persona, (3) 想要多视角意见, (4) 准备启动 persona-debate 但不知道叫谁来。 Triggers: "which persona", "ask multiple personas", "推荐 persona", "调度 persona", "哪个 persona", "问谁合适"
Plugin manifests1
{
"name": "persona-distill",
"description": "Persona Distillation Ecosystem — a parameterized schema library + generation/evaluation/scheduling/debate toolkit for distilling personas (people or rule systems) into self-contained Claude Code skills. 5 skills, 9 persona schemas, 19 reusable components (incl. execution-profile via Klein-RPD/CDM 4-sweep), 9-phase pipeline with v0.4.0 security hardening (consent attestation, untrusted-corpus discipline, fingerprint verification, self-containment linter, rubric config lock, corpus access declaration).",
"version": "0.4.0",
"author": {
"name": "XRenSiu",
"email": "[email protected]"
},
"homepage": "https://github.com/XRenSiu/claude-code-forge/tree/main/plugins/persona-distill",
"repository": "https://github.com/XRenSiu/claude-code-forge",
"license": "MIT",
"keywords": [
"persona-distillation",
"skill-generation",
"schema-library",
"quality-evaluation",
"multi-modal-collection",
"persona-routing",
"persona-debate",
"cognitive-os",
"self-contained-skills",
"execution-profile",
"cognitive-task-analysis",
"recognition-primed-decision"
]
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[persona-distill on Agent Plugins Marketplace](https://pluginsmp.com/plugins/persona-distill)