Skip to content
Back to skills

Lindy Multi Env Setup

ASecurity

Configure Lindy AI across development, staging, and production environments. Use when setting up multi-environment deployments, configuring per-environment secrets, or implementing environment-specific Lindy configurations. Trigger with phrases like "lindy environments", "lindy staging", "lindy dev prod", "lindy environment setup", "lindy config by env".

  • 2,688 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added February 7, 2026
devopstypescriptbashnodeawsapici/cd

Works with

  • cli
  • api

Security analysis

A100/100

Scanned February 12, 2026

npx -y skills add jeremylongshore/claude-code-plugins-plus-skills --skill lindy-multi-env-setup --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Lindy Multi Env Setup?

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

Security grade badge for Lindy Multi Env Setup
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jeremylongshore-lindy-multi-env-setup/badge)](https://www.skillsdirectory.com/skills/jeremylongshore-lindy-multi-env-setup)

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: lindy-multi-env-setup
description: |
  Configure Lindy AI across development, staging, and production environments.
  Use when setting up multi-environment deployments, configuring per-environment secrets,
  or implementing environment-specific Lindy configurations.
  Trigger with phrases like "lindy environments", "lindy staging",
  "lindy dev prod", "lindy environment setup", "lindy config by env".
allowed-tools: Read, Write, Edit, Bash(aws:*), Bash(gcloud:*), Bash(vault:*)
version: 1.0.0
license: MIT
author: Jeremy Longshore <jeremy@intentsolutions.io>
---

# Lindy Multi Env Setup

## Overview
Configure Lindy AI across development, staging, and production environments.

## Prerequisites
- Separate Lindy API keys per environment
- Secret management solution (Vault, AWS Secrets Manager, etc.)
- CI/CD pipeline with environment variables
- Environment detection in application

## Instructions

### Step 1: Create Environment Configuration
```typescript
// config/lindy.ts
interface LindyConfig {
  apiKey: string;
  environment: 'development' | 'staging' | 'production';
  baseUrl?: string;
  timeout: number;
  retries: number;
}

const configs: Record<string, LindyConfig> = {
  development: {
    apiKey: process.env.LINDY_DEV_API_KEY!,
    environment: 'development',
    timeout: 60000,
    retries: 1,
  },
  staging: {
    apiKey: process.env.LINDY_STAGING_API_KEY!,
    environment: 'staging',
    timeout: 45000,
    retries: 2,
  },
  production: {
    apiKey: process.env.LINDY_PROD_API_KEY!,
    environment: 'production',
    timeout: 30000,
    retries: 3,
  },
};

export function getLindyConfig(): LindyConfig {
  const env = process.env.NODE_ENV || 'development';
  return configs[env] || configs.development;
}
```

### Step 2: Implement Environment Detection
```typescript
// lib/lindy-client.ts
import { Lindy } from '@lindy-ai/sdk';
import { getLindyConfig } from '../config/lindy';

let client: Lindy | null = null;

export function getLindyClient(): Lindy {
  if (!client) {
    const config = getLindyConfig();

    // Validate environment
    if (config.environment === 'production') {
      if (!config.apiKey.startsWith('lnd_prod_')) {
        throw new Error('Production requires production API key');
      }
    }

    client = new Lindy({
      apiKey: config.apiKey,
      timeout: config.timeout,
      retries: config.retries,
    });
  }

  return client;
}
```

### Step 3: Configure Secrets by Environment
```yaml
# AWS Secrets Manager structure
secrets/
├── lindy/development
│   └── api_key: lnd_dev_xxx
├── lindy/staging
│   └── api_key: lnd_stg_xxx
└── lindy/production
    └── api_key: lnd_prod_xxx
```

```typescript
// secrets/lindy.ts
import { SecretsManager } from '@aws-sdk/client-secrets-manager';

export async function getLindyApiKey(env: string): Promise<string> {
  const client = new SecretsManager({ region: 'us-east-1' });

  const response = await client.getSecretValue({
    SecretId: `lindy/${env}`,
  });

  const secret = JSON.parse(response.SecretString!);
  return secret.api_key;
}
```

### Step 4: Environment-Specific Agents
```typescript
// agents/config.ts
interface AgentMapping {
  development: string;
  staging: string;
  production: string;
}

const agentMappings: Record<string, AgentMapping> = {
  support: {
    development: 'agt_dev_support',
    staging: 'agt_stg_support',
    production: 'agt_prod_support',
  },
  sales: {
    development: 'agt_dev_sales',
    staging: 'agt_stg_sales',
    production: 'agt_prod_sales',
  },
};

export function getAgentId(agentName: string): string {
  const env = process.env.NODE_ENV || 'development';
  const mapping = agentMappings[agentName];

  if (!mapping) {
    throw new Error(`Unknown agent: ${agentName}`);
  }

  return mapping[env as keyof AgentMapping];
}
```

### Step 5: Add Environment Guards
```typescript
// guards/production.ts
export function requireProduction(): void {
  if (process.env.NODE_ENV !== 'production') {
    throw new Error('This operation requires production environment');
  }
}

export function preventProduction(): void {
  if (process.env.NODE_ENV === 'production') {
    throw new Error('This operation is not allowed in production');
  }
}

// Usage
async function dangerousOperation() {
  preventProduction();
  // ... destructive test operation
}

async function productionOnlyOperation() {
  requireProduction();
  // ... production-only logic
}
```

## Output
- Multi-environment configuration
- Environment detection logic
- Secure secret management
- Environment-specific agents
- Production safeguards

## Error Handling
| Issue | Cause | Solution |
|-------|-------|----------|
| Wrong key for env | Config error | Validate key prefix |
| Secret not found | Not provisioned | Create in secrets manager |
| Agent not found | Wrong environment | Check agent mapping |

## Examples

### Complete Environment Setup
```typescript
// index.ts
import { getLindyClient } from './lib/lindy-client';
import { getAgentId } from './agents/config';

async function main() {
  const lindy = getLindyClient();
  const agentId = getAgentId('support');

  console.log(`Environment: ${process.env.NODE_ENV}`);
  console.log(`Agent: ${agentId}`);

  const result = await lindy.agents.run(agentId, {
    input: 'Test message',
  });

  console.log('Response:', result.output);
}

main().catch(console.error);
```

## Resources
- [Lindy Environments Guide](https://docs.lindy.ai/environments)
- [12-Factor App Config](https://12factor.net/config)
- [AWS Secrets Manager](https://aws.amazon.com/secrets-manager/)

## Next Steps
Proceed to `lindy-observability` for monitoring setup.

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.

Comments

Loading comments…

Newsletter

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