Agent skill
new-agent-creation
Provides step-by-step templates and guidance for creating new AI agents in Unite-Hub with proper registration, testing, and governance
Install this agent skill to your Project
npx add-skill https://github.com/aiskillstore/marketplace/tree/main/skills/cleanexpo/new-agent-creation
SKILL.md
New Agent Creation Skill
Step-by-Step Agent Implementation
When to Use: Creating new AI agents for Unite-Hub
Quick Start Template
1. Create Agent File
Location: src/lib/agents/my-new-agent.ts
import { BaseAgent, AgentTask, AgentConfig } from './base-agent';
import { getAnthropicClient } from '@/lib/anthropic/lazy-client';
export class MyNewAgent extends BaseAgent {
constructor() {
super({
name: 'MyNewAgent',
queueName: 'my-new-agent-queue',
concurrency: 1,
retryDelay: 5000
});
}
protected async processTask(task: AgentTask): Promise<any> {
// Your agent logic here
const client = getAnthropicClient();
const response = await client.messages.create({
model: 'claude-sonnet-4-5-20250929',
max_tokens: 4096,
messages: [{
role: 'user',
content: `Task: ${task.task_type}\nPayload: ${JSON.stringify(task.payload)}`
}]
});
return {
result: response.content[0].text,
workspace_id: task.workspace_id,
model_used: 'claude-sonnet-4-5-20250929',
input_tokens: response.usage.input_tokens,
output_tokens: response.usage.output_tokens
};
}
}
2. Register in Orchestrator
File: src/lib/agents/orchestrator-router.ts
// Add to AgentIntent enum
export type AgentIntent =
| 'my_new_agent' // ← Add this
| 'email' | 'content' | ...;
// Add to classifyIntent function
if (objective.includes('my task keyword')) {
return 'my_new_agent';
}
3. Add to Registry
File: .claude/agents/registry.json
{
"id": "my-new-agent",
"name": "My New Agent",
"version": "1.0.0",
"file": "src/lib/agents/my-new-agent.ts",
"capabilities": ["capability_1", "capability_2"],
"queueName": "my-new-agent-queue",
"models": ["claude-sonnet-4-5-20250929"],
"governance": "HUMAN_GOVERNED",
"verification_required": true,
"budget_daily_usd": 10.00,
"category": "marketing"
}
4. Create Tests
File: tests/agents/my-new-agent.test.ts
import { describe, it, expect, vi } from 'vitest';
import { MyNewAgent } from '@/lib/agents/my-new-agent';
describe('MyNewAgent', () => {
it('processes task successfully', async () => {
const agent = new MyNewAgent();
const task = {
id: 'test-1',
workspace_id: 'ws-123',
task_type: 'my_task',
payload: { data: 'test' },
priority: 5,
retry_count: 0,
max_retries: 3
};
const result = await agent.processTask(task);
expect(result).toBeDefined();
expect(result.workspace_id).toBe('ws-123');
});
});
5. Create CLI Runner (Optional)
File: scripts/run-my-agent.mjs
import { MyNewAgent } from '../src/lib/agents/my-new-agent.js';
const agent = new MyNewAgent();
await agent.start();
console.log('MyNewAgent running...');
Checklist
- Create agent file extending BaseAgent
- Implement processTask() method
- Add to orchestrator-router.ts intent enum
- Add to orchestrator routing logic
- Register in .claude/agents/registry.json
- Create tests (100% pass required)
- Add CLI runner script (optional)
- Document in .claude/agent.md
- Set appropriate budget limits
- Choose governance mode (HUMAN_GOVERNED vs AUTONOMOUS)
Standard: All agents must filter by workspace_id, respect budgets, pass verification
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
perigon-backend
Perigon ASP.NET Core + EF Core + Aspire conventions
perigon-agent
Pointers for Copilot/agents to apply Perigon conventions
perigon-angular
Angular 21+ standalone/Material/signal conventions for Perigon WebApp
fastapi-mastery
Comprehensive FastAPI development skill covering REST API creation, routing, request/response handling, validation, authentication, database integration, middleware, and deployment. Use when working with FastAPI projects, building APIs, implementing CRUD operations, setting up authentication/authorization, integrating databases (SQL/NoSQL), adding middleware, handling WebSockets, or deploying FastAPI applications. Triggered by requests involving .py files with FastAPI code, API endpoint creation, Pydantic models, or FastAPI-specific features.
context7-efficient
Token-efficient library documentation fetcher using Context7 MCP with 86.8% token savings through intelligent shell pipeline filtering. Fetches code examples, API references, and best practices for JavaScript, Python, Go, Rust, and other libraries. Use when users ask about library documentation, need code examples, want API usage patterns, are learning a new framework, need syntax reference, or troubleshooting with library-specific information. Triggers include questions like "Show me React hooks", "How do I use Prisma", "What's the Next.js routing syntax", or any request for library/framework documentation.
browser-use
Browser automation using Playwright MCP. Navigate websites, fill forms, click elements, take screenshots, and extract data. Use when tasks require web browsing, form submission, web scraping, UI testing, or any browser interaction.
Didn't find tool you were looking for?