LeoYeAI

Distill a teammate into an AI Skill. Auto-collect Slack/Teams/GitHub data, generate Work Skill + 5-layer Persona, with continuous evolution. Powered by MyClaw.ai. Works with Claude Code, OpenClaw, and any AgentSkills-compatible agent.

90
0
100% credibility
Found Apr 02, 2026 at 90 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This open-source tool creates AI personas mimicking departed teammates by processing their past messages, documents, and work artifacts into reusable skills for AI assistants.

How It Works

1
๐Ÿ˜ข A teammate leaves

Your helpful coworker departs, taking years of know-how with them, leaving you wishing their advice was still around.

2
๐Ÿ” Find teammate.skill

You discover this tool that turns their old chats, notes, and work into a living digital version of them.

3
๐Ÿ“ฅ Add to your AI helper

You easily place it into your everyday AI assistant, ready to bring back your teammate in moments.

4
๐Ÿง‘โ€๐Ÿ’ป Build your digital teammate

Describe who they were, share their old messages or files if you have them, and watch a preview come alive just like them.

5
๐Ÿ’ฌ Chat like old times

Ask questions or share code, and get back their exact style, opinions, and expertise every time.

6
๐Ÿ”„ Keep them growing

Add more stories or correct details anytime, so your digital teammate gets even sharper over time.

โœจ Team knowledge lives on

Now their wisdom is always there, helping your team forever, no more lost expertise.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 90 to 90 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is teammate-skill?

Teammate-skill turns departed teammates into persistent AI agents by auto-collecting their Slack, Teams, and GitHub data to distill a Work Skill plus 5-layer Persona that mimics their code style, review habits, and voice. Built in Python, it integrates with Claude Code, OpenClaw, or any AgentSkills-compatible agentโ€”run /create-teammate, feed in data or a description, and invoke via /{slug} for realistic advice. Solves tribal knowledge loss without endless docs.

Why is it gaining traction?

Beyond basic LLM distillation like deepseek distill github or distill llm github tools, it offers continuous evolution: append data, correct via chat, rollback versions, or compare teammates side-by-side with scenario sims. Auto-collect handles real data flows, while a built-in quality pipeline and privacy guard ensure usable, safe outputs. Devs dig the low-friction commands and portable exports for sharing agent skills.

Who should use this?

Engineering leads preserving senior expertise after exits, via teammate skills that review PRs or debate MongoDB vs Postgres. Backend teams wanting Stripe-style critics or Google L5 personas for design critiques. Anyone on Claude Code or OpenClaw practicing with 5-layer agents from teammate practice and skills book ideas, without manual distill web monitor github setups.

Verdict

Worth cloning for Claude/OpenClaw usersโ€”90 stars and 1.0% credibility reflect early maturity, but multilingual docs, privacy scans, and MIT license lower risks. Prototype one teammate skill; it'll evolve faster than static alternatives.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.