Skip to content
Back to skills

Boss Skills

ASecurity

Distill a real boss into an AI skill, or generate a boss skill from famous entrepreneur archetypes (Elon Musk, Steve Jobs, Jeff Bezos, Jensen Huang). Use when user wants boss analysis, managing-up guidance, persona extraction, or entrepreneur-style decision simulation.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agentspythongobashgit

Security analysis

A100/100

Pro scans all 20 files and shows the line behind each finding

Scanned November 11, 2026

npx -y skills add YYC-Cube/YYC3-AI-Agent-Archive --skill boss-skills --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Boss Skills?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Boss Skills
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yyc-cube-boss-skills/badge)](https://www.skillsdirectory.com/skills/yyc-cube-boss-skills)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: boss-skills
description: "Distill a real boss into an AI skill, or generate a boss skill from famous entrepreneur archetypes (Elon Musk, Steve Jobs, Jeff Bezos, Jensen Huang). Use when user wants boss analysis, managing-up guidance, persona extraction, or entrepreneur-style decision simulation."
category: development-code
description_zh: "蒸馏老板或生成企业家原型 Skill,模拟管理风格与决策模式"
description_en: "Distill bosses or generate entrepreneur archetype Skills for management simulation"
version: "1.1.0"
homepage: https://github.com/vogtsw/boss-skills
allowed-tools: Read, Write, Edit, Bash
---

# Create Boss

Use this skill in two modes:

1. `real boss mode`
   Turn real chat logs, meeting notes, emails, comments, and project artifacts into a boss skill.
2. `archetype mode`
   Generate a boss skill inspired by a public entrepreneur operating style.

## Trigger phrases

- `/create-boss`
- `/list-bosses`
- `/boss-rollback`
- `/delete-boss`
- "create a boss skill"
- "analyze my boss"
- "build a Musk-style boss"
- "make a Steve Jobs style leader"
- "give me a Bezos-style management model"
- "list boss archetypes"

## Tools

- Parse imported material with the files in [`tools/`](tools).
- Write or update generated boss skills with [`tools/skill_writer.py`](tools/skill_writer.py).
- Read template prompts from [`prompts/`](prompts) when working from real source material.
- Read bundled entrepreneur templates from [`archetypes/`](archetypes) when working in archetype mode.

These scripts are internal implementation details for the agent.
Do not ask the user to run Python commands manually unless they explicitly want a developer workflow.

## Workflow

### Mode 1: Real Boss

1. Ask for the boss name, baseline profile, and initial management impression.
2. Ask for source material: chats, meeting notes, docs, email, or pasted text.
3. Distill three outputs:
   - `judgment.md`
   - `management.md`
   - `persona.md`
4. Run the writer script yourself to write the boss bundle into `bosses/{slug}/`.
5. Show the generated commands:
   - `/{slug}`
   - `/{slug}-judgment`
   - `/{slug}-management`
   - `/{slug}-persona`

### Mode 2: Entrepreneur Archetype

1. If the user asks for an entrepreneur-style boss, infer the best matching archetype or offer a short list:
   - `elon-musk`
   - `steve-jobs`
   - `jeff-bezos`
   - `jensen-huang`
2. Run the writer script yourself to generate the skill. Do not expose the internal command as the primary UX.
3. Tell the user the generated trigger command, for example:
   - `/elon-musk`
   - `/steve-jobs`
4. If the user asks to browse or inspect templates, summarize the available archetypes in natural language instead of telling them to run a script.

## Management Commands

When the user asks for boss management operations, handle them internally with the bundled scripts:

- `/list-bosses`
  Run `tools/skill_writer.py --action list` and summarize the available boss skills.
- `/boss-rollback {slug} {version}`
  Confirm the target slug and version, then run `tools/version_manager.py --action rollback`.
- `/delete-boss {slug}`
  Confirm before deletion, then run `tools/skill_writer.py --action delete --slug {slug}`.

Do not tell normal users to copy these commands manually. Execute the workflow yourself and report the result.

## Bundled Archetypes

- `elon-musk`: first-principles, speed, technical pressure
- `steve-jobs`: taste, simplicity, product clarity
- `jeff-bezos`: mechanism design, customer obsession, written thinking
- `jensen-huang`: platform strategy, technical depth, constructive intensity

## Files Created

Every generated boss skill should include:

- `SKILL.md`
- `judgment.md`
- `management.md`
- `persona.md`
- `meta.json`
- `judgment_skill.md`
- `management_skill.md`
- `persona_skill.md`

## Safety Framing

- Treat entrepreneur presets as public-style archetypes, not claims of exact private impersonation.
- Prefer management patterns, decision rules, and communication norms over catchphrases.
- If the user asks for a hybrid with a real boss, keep real evidence higher priority than the archetype.

Files in this skill

  • SKILL.md4 KB
  • agents/openai.yaml400 B
  • archetypes/elon-musk/judgment.md1.4 KB
  • archetypes/elon-musk/management.md1.1 KB
  • archetypes/elon-musk/meta.json813 B
  • archetypes/elon-musk/persona.md881 B
  • archetypes/jeff-bezos/judgment.md1.3 KB
  • archetypes/jeff-bezos/management.md1 KB
  • archetypes/jeff-bezos/meta.json798 B
  • archetypes/jeff-bezos/persona.md881 B
  • archetypes/jensen-huang/judgment.md1.4 KB
  • archetypes/jensen-huang/management.md1.1 KB
  • archetypes/jensen-huang/meta.json812 B
  • archetypes/jensen-huang/persona.md883 B
  • archetypes/steve-jobs/judgment.md1.3 KB
  • archetypes/steve-jobs/management.md987 B
  • archetypes/steve-jobs/meta.json791 B
  • archetypes/steve-jobs/persona.md847 B
  • bosses/example-laozhou/SKILL.md6.3 KB
  • bosses/example-laozhou/judgment.md1.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.