research-app-toolkit
Research application toolkit for PhD, MRes, MPhil applications - CV analysis, professor matching, cold email optimization, document assistance, school selection, interview prep, interview PPT preparation, and research gap analysis
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/xujingchen1996--research-app-toolkit)HTML
<a href="/listings/xujingchen1996--research-app-toolkit"><img src="/api/listings/xujingchen1996--research-app-toolkit/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Recommended: npm CLI
npm install -g @xujingchen1996/research-app-toolkit ratk install
Optional host selection:
ratk install --claude ratk install --codex ratk install --all
Default behavior:
- `ratk install` installs to the hosts detected on the current machine - `ratk install --claude` installs only the Claude Code plugin - `ratk install --codex` installs only the Codex plugin - `ratk install --all` installs to every detected host and skips hosts that are not installed
You can also run the installer without a global install:
npx @xujingchen1996/research-app-toolkit install
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/plugin.json · Manifest · 12 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- {
- "name": "research-app-toolkit",
- "version": "0.3.0",
- "description": "Research application toolkit for PhD, MRes, MPhil applications - CV analysis, professor matching, cold email optimization, document assistance, school selection, interview prep, interview PPT preparation, and research gap analysis",
- "author": {
- "name": "Jingchen Xu"
- },
- "license": "MIT",
- "keywords": ["phd", "mres", "mphil", "application", "cv", "cold-email", "professor", "research", "university", "interview", "ppt", "presentation"],
- "commands": "./commands/",
- "skills": "./skills/"
- }
README.md · README · 159 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # Research Application Toolkit
- [](https://github.com/xujingchen1996/research-app-toolkit/stargazers)
- [](https://github.com/xujingchen1996/research-app-toolkit/releases)
- [](./LICENSE)
- [](https://claude.ai/code)
- [](https://openai.com)
- Research Application Toolkit is a single repository that ships host-specific integrations for both Claude Code and Codex.
- It covers:
- - CV analysis and profile extraction
- - CV refinement for target programs
- - Supervisor matching
- - Outreach and follow-up emails
- - SOP, personal statement, motivation letter, and proposal drafting
- - School selection
- - Interview preparation
- - Interview PPT / professor meeting deck preparation
- - Research gap analysis
- ## Install
- ### Recommended: npm CLI
- ```bash
- npm install -g @xujingchen1996/research-app-toolkit
- ratk install
- ```
- Optional host selection:
- ```bash
- ratk install --claude
- ratk install --codex
- ratk install --all
- ```
- Default behavior:
- - `ratk install` installs to the hosts detected on the current machine
- - `ratk install --claude` installs only the Claude Code plugin
- - `ratk install --codex` installs only the Codex plugin
- - `ratk install --all` installs to every detected host and skips hosts that are not installed
- You can also run the installer without a global install:
- ```bash
- npx @xujingchen1996/research-app-toolkit install
- ```
- ## Usage
- ### Claude Code
- After installation into Claude Code, use the slash commands:
- ```bash
- /ra:cv-analyze
- /ra:cv-polish
- /ra:professor-match
- /ra:cold-email
- /ra:doc-assist
- /ra:school-select
- /ra:interview-prep
- /ra:interview-ppt
- /ra:gap-analysis
- ```
- ### Codex
- After installation into Codex, use the plugin in chat with prompts such as:
- ```text
- Analyze my CV and extract a research application profile
- Match supervisors and schools for my target research area
- Draft a bilingual outreach email to a potential supervisor
- Help me prepare a professor meeting or PhD interview PPT
- ```
- The Codex host keeps its own `codex/memory.md` state and supports `zh`, `en`, and `bilingual` output modes.
- ## Repository Layout
- ```text
- research-app-toolkit/
- ├── .claude-plugin/ # Claude Code manifest
- ├── .codex-plugin/ # Codex manifest
- ├── .local.md # Claude Code shared state template
- ├── assets/ # Shared assets such as the Codex icon
- ├── commands/ # Claude Code slash commands
- ├── hooks/ # Claude Code hooks
- ├── agents/ # Claude Code helper agents
- ├── skills/ # Claude Code skills
- ├── codex/
- │ ├── memory.md # Codex shared state
- │ └── skills/ # Codex skills and bundle manifest
- ├── bin/ # npm CLI entrypoint
- ├── lib/ # Installer helpers
- ├── test/ # Installer tests
- ├── package.json
- └── README.md
- ```
- ## Development Notes
- - Claude Code and Codex skills are intentionally maintained separately to avoid host-specific prompt interference.
- - Shared installation is handled by the `ratk` CLI rather than host auto-discovery from a cloned repository.
- - Local host state such as `~/.claude`, `~/.codex`, `~/.agents`, and Codex caches should not be committed to this repository.
- ## Test
- ```bash
- npm test
- ```
- ## License
- MIT
- ## DeepSeek Harness(DSH)
- 同一个 npm 包提供 RA 科研申请预设、九个技能,以及申请档案、选校与导师、邮件与文书、面试准备四个 UI 标签页。DSH 当前验证版本为 0.2.0-rc.2。
- ### 尚未安装 DSH:先准备文件
- ```bash
- npm install -g @xujingchen1996/research-app-toolkit
- ratk install --dsh
- ```
- 也可直接运行:
- ```bash
- npx @xujingchen1996/research-app-toolkit install --dsh
- ```
- 此步骤不需要 DSH,也不会自动安装 DSH 或修改 profile。它将 bundle 放到 `~/.dsh/bundles/research-app-toolkit/`;若设置 `DSH_HOME`,则使用 `$DSH_HOME/bundles/research-app-toolkit/`。预安装成功表示文件已就绪,使用预设和 UI 仍需要安装 DSH。
- ### 已安装 DSH:启用预设与 UI
- 1. 打开 DSH 的“插件 → 添加插件”。
- 2. 输入安装器输出的完整本地目录,或者包名 `@xujingchen1996/research-app-toolkit`。
- 3. 安装后点击“立即启用”,在新会话选择“RA 科研申请模式”。
- 4. Web 与 Desktop 的 profile 独立,分别添加并启用。旧会话若持有旧工具版本,重启宿主或新建 RA 会话。
- 如果已手工安装旧的 `@local/dsh-preset-research-app-toolkit` 与 `dsh-ra-studio`,先在插件管理器停用这两个 bundle,再启用新包,避免重复预设/接口。此操作不删除工作区档案。
- ### 使用与数据
- - 九个技能:`/cv-analyze`、`/cv-polish`、`/professor-match`、`/school-select`、`/cold-email`、`/document-assist`、`/gap-analysis`、`/interview-prep`、`/interview-ppt`。
- - 申请档案:当前会话工作区 `.ra/memory.md`;工作台记录:`.ra/workbench.json`。模板不会覆盖已有个人材料。
- - UI 支持编辑、保存和准备任务;“准备任务”只填写对话草稿,用户发送后模型才处理。模型与 UI 通过 `ra_workbench` 共用数据。
- - Zotero、图像生成、生命科学专用技能和 PPTX 导出工具属于可选能力,包不会自动安装它们。
- 普通安装与测试不需要 DSH;本包没有运行时依赖,DSH 仅声明为可选 peer dependency。
- 发布前检查:`npm run check`、`npm test`。已安装 DSH 时再运行 `npm run test:dsh`,该检查可用 `DSH_NODE_MODULES` 指定 DSH 的 node_modules 目录。
Version history
v0.3.010/19/2026
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.