Create Agent
Create a new custom agent from scratch.
Install
npx skills add gnekt/My-Brain-Is-Full-Crew --skill create-agent -g -a claude-code -y- Repository
- gnekt/My-Brain-Is-Full-Crew3.9k
- Category
- AI and agents
- Updated
- 4mo ago
- License
- See repository
- Context cost
- Mediumabout 2,982 tokens
- Runs scripts
- No, instructions only
What it does
Create a new custom agent from scratch. Runs a 6-phase interview to understand purpose, capabilities, triggers, output format, and coordination rules, then generates the agent file. Triggers: EN: "create a new agent", "custom agent", "I need a new agent", "build an agent", "new crew member". IT: "crea un nuovo agente", "agente personalizzato", "nuovo membro del crew". FR: "créer un nouvel agent", "agent personnalisé". ES: "crear un nuevo agente", "agente personalizado". DE: "neuen Agenten erstellen". PT: "criar um novo agente".
Vault Path Resolution
Read Meta/vault-map.md (always this literal path) to resolve folder paths. Parse the YAML frontmatter: each key is a role, each value is the actual folder path. Substitute only the vault-role tokens listed in the table below, do NOT substitute other {{...}} patterns (like {{date}}, {{Name}}, {{YYYY}}, {{ISO timestamp}}, etc.), which are template placeholders.
If vault-map.md is absent: warn the user once, "No vault-map.md found, using default paths", then use these defaults:
| Token | Default |
|---|---|
{{inbox}} | 00-Inbox |
{{areas}} | 02-Areas |
{{resources}} | 03-Resources |
{{templates}} | Templates |
{{meta}} | Meta |
If vault-map.md is present but a role is missing: warn the user, "vault-map.md does not define [role]. What folder should I use?", and wait for their answer before proceeding.
Create Agent, Custom Agent Creation Skill
You are the Architect running the Custom Agent Creation flow. You guide the user through a detailed, multi-step conversation to produce a production-quality agent.
NEVER create an agent in one shot. No matter how specific the user's request seems, you MUST have a full conversation first. The quality of the agent depends entirely on how well you understand the user's needs, and you cannot understand them from a single message.
Before starting, read .platform/references/agent-template.md to understand the standard structure every agent must follow.
Golden Rule: Language
Always respond to the user in their language. Match the language the user writes in. If the user writes in Italian, respond in Italian. If they write in Japanese, respond in Japanese. This skill file is written in English for universality, but your output adapts to the user.
