agent-creator

Agent Creator

$ARGUMENTS

Create a new specialized agent following ai-toolkit conventions.

Workflow

1. Capture role -- what problem space should the agent own?
2. Define triggers -- which keywords or task types should route to this agent?
3. Choose tools -- minimal tool set, least privilege first
4. Choose model -- opus for deep reasoning, sonnet for lighter pattern work
5. Map supporting skills -- which knowledge skills should the agent reference?
6. Write instructions -- capabilities, constraints, escalation rules, deliverables
7. Validate -- frontmatter, naming, skills references, tool whitelist

Required Frontmatter

---
name: agent-name
description: "When to use this agent. Triggers: keyword1, keyword2."
tools: Read, Write, Edit
model: sonnet
skills: skill-one, skill-two
---

Authoring Rules

• MUST match filename and name: using lowercase-hyphen format — drift breaks routing
• MUST include an explicit Triggers: list in the description so the router can dispatch deterministically
• NEVER reference skills that do not exist — either create the dependency first or drop the reference
• NEVER grant write access to .claude/agents/ — that authority belongs to meta-architect alone
• CRITICAL: avoid tool bloat. Every extra tool widens blast radius; start from Read and justify additions one by one
• Give the agent a clear boundary: what it owns and what it must escalate
• Prefer specialized, narrow responsibility over generic "do everything" agents

Agent Skeleton

---
name: {agent-name}
description: "When to use this agent. Triggers: keyword1, keyword2."
tools: Read, Edit
model: sonnet
skills: relevant-skill
---

You are a specialized agent for {domain}.

## Responsibilities
- Responsibility one
- Responsibility two

## Constraints
- Do not edit files outside owned scope unless required
- Escalate security, data-loss, or architecture risks

## Deliverables
- Clear findings
- Specific edits or recommendations
- Validation notes