home / skills / jeremylongshore / claude-code-plugins-plus-skills / clay-sdk-patterns

This skill helps you apply production-ready Clay SDK patterns in TypeScript and Python to improve reliability and consistency.

npx playbooks add skill jeremylongshore/claude-code-plugins-plus-skills --skill clay-sdk-patterns

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
3.5 KB
---
name: clay-sdk-patterns
description: |
  Apply production-ready Clay SDK patterns for TypeScript and Python.
  Use when implementing Clay integrations, refactoring SDK usage,
  or establishing team coding standards for Clay.
  Trigger with phrases like "clay SDK patterns", "clay best practices",
  "clay code patterns", "idiomatic clay".
allowed-tools: Read, Write, Edit
version: 1.0.0
license: MIT
author: Jeremy Longshore <[email protected]>
---

# Clay SDK Patterns

## Overview
Production-ready patterns for Clay SDK usage in TypeScript and Python.

## Prerequisites
- Completed `clay-install-auth` setup
- Familiarity with async/await patterns
- Understanding of error handling best practices

## Instructions

### Step 1: Implement Singleton Pattern (Recommended)
```typescript
// src/clay/client.ts
import { ClayClient } from '@clay/sdk';

let instance: ClayClient | null = null;

export function getClayClient(): ClayClient {
  if (!instance) {
    instance = new ClayClient({
      apiKey: process.env.CLAY_API_KEY!,
      // Additional options
    });
  }
  return instance;
}
```

### Step 2: Add Error Handling Wrapper
```typescript
import { ClayError } from '@clay/sdk';

async function safeClayCall<T>(
  operation: () => Promise<T>
): Promise<{ data: T | null; error: Error | null }> {
  try {
    const data = await operation();
    return { data, error: null };
  } catch (err) {
    if (err instanceof ClayError) {
      console.error({
        code: err.code,
        message: err.message,
      });
    }
    return { data: null, error: err as Error };
  }
}
```

### Step 3: Implement Retry Logic
```typescript
async function withRetry<T>(
  operation: () => Promise<T>,
  maxRetries = 3,
  backoffMs = 1000
): Promise<T> {
  for (let attempt = 1; attempt <= maxRetries; attempt++) {
    try {
      return await operation();
    } catch (err) {
      if (attempt === maxRetries) throw err;
      const delay = backoffMs * Math.pow(2, attempt - 1);
      await new Promise(r => setTimeout(r, delay));
    }
  }
  throw new Error('Unreachable');
}
```

## Output
- Type-safe client singleton
- Robust error handling with structured logging
- Automatic retry with exponential backoff
- Runtime validation for API responses

## Error Handling
| Pattern | Use Case | Benefit |
|---------|----------|---------|
| Safe wrapper | All API calls | Prevents uncaught exceptions |
| Retry logic | Transient failures | Improves reliability |
| Type guards | Response validation | Catches API changes |
| Logging | All operations | Debugging and monitoring |

## Examples

### Factory Pattern (Multi-tenant)
```typescript
const clients = new Map<string, ClayClient>();

export function getClientForTenant(tenantId: string): ClayClient {
  if (!clients.has(tenantId)) {
    const apiKey = getTenantApiKey(tenantId);
    clients.set(tenantId, new ClayClient({ apiKey }));
  }
  return clients.get(tenantId)!;
}
```

### Python Context Manager
```python
from contextlib import asynccontextmanager
from clay import ClayClient

@asynccontextmanager
async def get_clay_client():
    client = ClayClient()
    try:
        yield client
    finally:
        await client.close()
```

### Zod Validation
```typescript
import { z } from 'zod';

const clayResponseSchema = z.object({
  id: z.string(),
  status: z.enum(['active', 'inactive']),
  createdAt: z.string().datetime(),
});
```

## Resources
- [Clay SDK Reference](https://docs.clay.com/sdk)
- [Clay API Types](https://docs.clay.com/types)
- [Zod Documentation](https://zod.dev/)

## Next Steps
Apply patterns in `clay-core-workflow-a` for real-world usage.

Overview

This skill provides production-ready Clay SDK patterns for TypeScript and Python to build reliable, maintainable integrations. It bundles recommended patterns like a singleton/factory client, safe error wrappers, retry with exponential backoff, and runtime validation. Use it to standardize SDK usage across projects and teams.

How this skill works

The skill inspects common integration points and shows concrete code patterns: a singleton client for single-tenant apps, a factory for multi-tenant setups, async context managers for Python, and Zod schemas for response validation. It also supplies reusable helpers: a safe call wrapper that catches and logs Clay errors, and a withRetry helper that retries transient failures with exponential backoff. Combine these patterns to get type-safe clients, structured logging, and resilient API calls.

When to use it

  • Implementing a new Clay integration in TypeScript or Python
  • Refactoring existing SDK usage to improve reliability and observability
  • Establishing team-level coding standards for Clay API calls
  • Handling transient network or API errors with retries
  • Validating remote API responses at runtime

Best practices

  • Create a singleton or tenant-factory to centralize client config and reuse connections
  • Wrap all SDK calls with a safe wrapper that returns {data, error} and logs structured errors
  • Use exponential backoff retries for idempotent operations prone to transient failures
  • Validate API responses with runtime schemas (Zod or type guards) to detect breaking changes early
  • Close clients or use async context managers in Python to avoid resource leaks

Example use cases

  • Single-tenant web server: getClayClient() singleton to share the client across requests
  • Multi-tenant platform: tenant client factory keyed by tenant ID with per-tenant API keys
  • Background job processor: safeClayCall + withRetry to process webhook retries reliably
  • CLI or script: runtime validation of responses to fail fast on schema drift
  • Testing: inject a mocked ClayClient implementation into the singleton or factory for deterministic tests

FAQ

How do I choose between a singleton and a factory?

Use a singleton for single-tenant apps or simple services. Use a factory map when each tenant or workspace needs its own API key or configuration.

When should I retry vs. fail fast?

Retry for transient errors (timeouts, 5xx). Fail fast for client errors (4xx) or when operations are non-idempotent unless you implement safe deduplication.