Skip to content
Back to search

ai-profile

SkillWorks with CodexGitHub

AI 能力画像:让你常用的任何 AI 根据真实使用记录,生成 30 项能力评估报告(雷达图 + 证据 + 改进建议)。Claude Code、Codex 等装 skill 即用;DeepSeek、千问、WorkBuddy 等网页版发文档即可。| AI Profile: any AI you use assesses 30 AI-usage skills from your real history. Install as a skill, or send the doc to any chat AI.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/billwonghk--ai-profile/badge.svg)](/listings/billwonghk--ai-profile)

HTML

<a href="/listings/billwonghk--ai-profile"><img src="/api/listings/billwonghk--ai-profile/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

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.

README.md · README · 127 lines
  1. # AI 能力画像 · AI Profile
  2. 让你最常用的 AI,根据你和它的真实使用记录,给你做一份 30 项的能力画像:六个领域、雷达图、每项的证据和改进建议。
  3. Let the AI you use most build a 30-item capability profile from your real usage history: six domains, radar charts, and evidence plus improvement tips for every item.
  4. 当前版本 / Current version:**0.7**
  5. ---
  6. ## 使用步骤 · How to use
  7. 按你用的 AI 工具,选下面对应的一种方式。
  8. Pick the method that matches the AI tool you use.
  9. ### 方式一:命令行安装 · Option 1: Install from the command line
  10. 适用于 Claude Code、Codex、Gemini CLI、Cursor、Antigravity 等支持 Agent Skills 标准的工具。
  11. For Claude Code, Codex, Gemini CLI, Cursor, Antigravity and other tools that support the Agent Skills standard.
  12. 1. 确认电脑上装了 Node.js。
  13. Make sure Node.js is installed.
  14. 2. 在终端运行:
  15. Run in your terminal:
  16. ```
  17. npx skills add billwonghk/ai-profile
  18. ```
  19. 3. 打开你的 AI 工具,说一句:“给我做一份 AI 能力画像”。
  20. Open your AI tool and say: "Build my AI profile".
  21. 4. 等它跑完,打开生成的 `ai-profile.html` 查看报告。
  22. When it finishes, open the generated `ai-profile.html` to view the report.
  23. 以后更新:`npx skills update ai-profile`
  24. To update later: `npx skills update ai-profile`
  25. ### 方式二:手动放入 skills 文件夹 · Option 2: Copy into the skills folder
  26. 适用于 WorkBuddy,以及不想用命令行的情况。
  27. For WorkBuddy, or if you'd rather not use the command line.
  28. 1. 下载 [`dist/ai-profile.zip`](dist/ai-profile.zip) 并解压,得到 `ai-profile` 文件夹。
  29. Download [`dist/ai-profile.zip`](dist/ai-profile.zip) and unzip it to get the `ai-profile` folder.
  30. 2. 把整个 `ai-profile` 文件夹放进你的工具的 skills 目录:
  31. Put the whole `ai-profile` folder into your tool's skills directory:
  32. | 工具 · Tool | 目录 · Directory |
  33. |---|---|
  34. | WorkBuddy(Mac) | `~/.workbuddy/skills/` |
  35. | WorkBuddy(Windows) | `%APPDATA%\WorkBuddy\skills\` |
  36. | Claude Code | `~/.claude/skills/` |
  37. | Codex | `~/.codex/skills/` |
  38. | Gemini CLI | `~/.gemini/skills/` |
  39. 3. 重启工具,在技能列表里能看到 ai-profile 就说明装好了。
  40. Restart the tool. If ai-profile appears in the skill list, it's installed.
  41. 4. 说一句:“给我做一份 AI 能力画像”,然后打开生成的 `ai-profile.html`。
  42. Say "Build my AI profile", then open the generated `ai-profile.html`.
  43. ### 方式三:Claude 网页版和 App · Option 3: Claude on the web and in the apps
  44. 1. 在 Settings → Capabilities 里,确认已经打开“代码执行和文件创建”(新版设置里叫 Create and edit files)。
  45. In Settings → Capabilities, make sure "Code execution and file creation" (called "Create and edit files" in the newer layout) is on.
  46. 2. 下载 [`dist/ai-profile.zip`](dist/ai-profile.zip),不用解压。
  47. Download [`dist/ai-profile.zip`](dist/ai-profile.zip). Don't unzip it.
  48. 3. 进入 Customize → Skills,点“+”,选 Create skill → Upload a skill,上传这个压缩包。
  49. Go to Customize → Skills, click "+", choose Create skill → Upload a skill, and upload the ZIP.
  50. 4. 新开一个对话,说:“给我做一份 AI 能力画像”。
  51. Start a new chat and say: "Build my AI profile".
  52. ### 方式四:网页版 AI · Option 4: Chat-only AIs
  53. 适用于 DeepSeek、千问、ChatGPT 网页版等不能安装 skill 的 AI。
  54. For AIs that can't install skills, such as DeepSeek, Qwen or ChatGPT on the web.
  55. 1. 下载 [`docs/AI能力画像-评估指令-v0.7.md`](docs/AI能力画像-评估指令-v0.7.md)。
  56. Download [`docs/AI能力画像-评估指令-v0.7.md`](docs/AI能力画像-评估指令-v0.7.md).
  57. 2. 把这个文件作为附件发给你的 AI,同时发送文档开头“给用户的说明”里的那段启动语。
  58. Send the file to your AI as an attachment, together with the start prompt in the "给用户的说明" section at the top of the document.
  59. 3. AI 会生成报告网页;生成不了网页的,会输出 Markdown 报告。
  60. The AI generates a report page, or a Markdown report if it can't make web pages.
  61. 4. 拿到 Markdown 报告时:下载本仓库的 [`ai-profile/assets/report.html`](ai-profile/assets/report.html),用浏览器打开,点右上角“粘贴报告”,就能看到图文版。
  62. If you get a Markdown report: download [`ai-profile/assets/report.html`](ai-profile/assets/report.html) from this repository, open it in a browser, click "粘贴报告" (Paste report) at the top right, and paste it in.
  63. ### 用了好几个 AI?· Using several AIs?
  64. 每个 AI 能看到的记录不一样,可以在常用的几个 AI 里各跑一次,再把几份报告一起发给其中一个 AI,说:“按评估规则第九步,合并这几份报告”。
  65. Each AI sees different records. Run the profile in each AI you use often, then send all the reports to one of them and say: "Merge these reports following step 9 of the rules."
  66. ---
  67. ## 报告里有什么 · What's in the report
  68. - 加权平均分、三到五句总评、最强的地方、最该补的五项
  69. Weighted average score, a short summary, top strengths, and the five things to improve first
  70. - 六个领域的雷达图:认知与判断、学习与执行、AI 驾驭力、业务与价值、沟通与协作、自我管理
  71. Radar charts for six domains: cognition, learning and execution, AI mastery, business value, communication, and self-management
  72. - 30 项逐项评分,每项附证据、反面证据和改进建议
  73. Scores for all 30 items, each with evidence, counter-evidence and an improvement tip
  74. - 使用数据:对话数、活跃天数、单一工具最长使用时长等
  75. Usage data: conversation counts, active days, longest use of a single tool, and more
  76. 报告用你平时和 AI 交流的语言来写。报告网页的界面目前是中文。网页右上角可以导出 Markdown 原报告。
  77. The report is written in the language you normally use with your AI. The report page interface is currently in Chinese. Use the button at the top right of the page to export the Markdown report.
  78. ## 说明 · Notes
  79. - AI 读取本机记录时只读、只统计,不修改任何文件。
  80. When reading local records, the AI only reads and counts. It never changes any file.
  81. - 这份评估由 AI 根据自己能看到的使用记录生成,不是标准化心理测评。
  82. The assessment is generated by an AI from the usage records it can see. It is not a standardized psychological test.
  83. - 只评能力,不评品行;不输出项目名、目录名、人名和个人敏感信息。
  84. It assesses ability only, not character, and never outputs project names, folder names, personal names or sensitive personal information.
  85. - 在任何一个 AI 上使用不足半年的,报告会提示参考意义有限。
  86. If you have used no single AI tool for at least six months, the report says its results are of limited value.
  87. ## 更新记录 · Changelog
  88. - **0.7**:30 项;新增“AI 使用资历”“优先级管理”“复盘与沉淀”;“成型与收敛”“价值与反馈”合并评估,“价值与反馈”权重减半;统一使用电脑系统时区;统一报告模板。
  89. 30 items; added "AI tenure", "Prioritization" and "Retrospection & reuse"; merged "Completion & focus" and "Value & feedback", with "Value & feedback" at half weight; all times use the computer's system time zone; one shared report template.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.