Colleague.skill

Distill a colleague into an AI Skill. Auto-collect Feishu/DingTalk data, generate Work Skill + Persona, with continuous evolution.

Category

People Role

Install

Hot:139

Download and extract to your skills directory

Copy command and send to OpenClaw for auto-install:

Download and install this skill https://openskills.cc/api/download?slug=titanwings-colleague-skill&locale=en&source=copy

Create-Colleague - Turn a Colleague into an AI Skill

Skill Overview


Create-colleague is a tool that transforms a colleague’s work capabilities and communication style into AI skills. It automatically collects data from Feishu/DingTalk to generate an evolvable skill containing both Work Skill (work ability) and Persona (personality traits).

Use Cases


  • Quick onboarding for new hires: New hires can use the AI skill to learn a senior’s code style, CR preferences, and technical decision logic, helping them integrate into the team faster.

  • Preserve knowledge from departing colleagues: Convert a departing colleague’s Feishu/DingTalk chat history and documents into interactive AI skills to retain key knowledge.

  • Seamless cross-timezone collaboration: Use the AI skill as a stand-in so colleagues in other time zones can provide feedback in your style.
  • Core Features


  • Automatic data collection: Supports automatic collection from Feishu group chats, private chats, documents, Wikis, and DingTalk documents/knowledge bases—no manual整理 required.

  • Dual-dimension modeling: Separately generates Work Skill (technical abilities, work processes, output standards) and Persona (communication style, decision-making patterns, interpersonal preferences) to accurately reproduce the colleague’s traits.

  • Continuous evolution: Supports adding new files and correcting through conversation. The skill keeps learning and improving—becoming more like them the more you use it.
  • FAQ

    What data sources does create-colleague support?


    It supports two major platforms: Feishu and DingTalk. Feishu can collect group chat messages, private chats, documents, and Wikis; DingTalk can collect documents and knowledge bases, while message history needs to be obtained via a browser. It also supports directly uploading PDFs, images, exported chat history files, emails, and other formats.

    How accurate are the generated skills?


    Accuracy depends on the amount and quality of the provided data. The richer the chat history and the more complete the documents, the more accurate the skill will be. The skills use a two-layer structure—Work + Persona—so they can reproduce not only technical abilities but also communication style. If something feels off, you can directly say, “He wouldn’t do it like that,” and the skill will continue learning and correcting.

    What’s the difference between Work Skill and Persona?


    Work Skill focuses on work capabilities, including the tech stack, systems ownership, CR priorities, code style, and more. Persona focuses on personality traits, including communication methods, decision-making patterns, emotional responses, and so on. The two parts can be called independently or used together.