agentsclimarketplace

Cohere multi env setup

Skill ComeOnOliver/skillshub/skills/jeremylongshore/claude-code-plugins-plus-skills/cohere-multi-env-setup

🧠 The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.

Install
npx -y skills add ComeOnOliver/skillshub --skill cohere-multi-env-setup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Configure Cohere across development, staging, and production environments. Use when setting up multi-environment deployments, configuring per-environment API keys, model selection, and rate limit strategies. Trigger with phrases like "cohere environments", "cohere staging", "cohere dev prod", "cohere environment setup", "cohere config by env".

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

7.3 KB, ~1.7k tokens by cl100k_base, as published. Nobody here has run it

Cohere Multi-Environment Setup

Overview

Configure Cohere API v2 across dev/staging/prod with environment-specific API keys, model selection, and budget controls.

Prerequisites

  • Separate Cohere API keys per environment (trial for dev, production for staging/prod)
  • Secret management solution (Vault, AWS Secrets Manager, GCP Secret Manager)
  • Environment detection in application

Environment Strategy

EnvironmentAPI Key TypeModelmaxTokensCaching
DevelopmentTrial (free)command-r7b-12-2024200Disabled
StagingProductioncommand-r-08-20241000Enabled
ProductionProductioncommand-a-03-20254096Enabled

Instructions

Step 1: Configuration Structure

// src/config/cohere.ts
type Environment = 'development' | 'staging' | 'production';

interface CohereEnvConfig {
  chatModel: string;
  embedModel: string;
  rerankModel: string;
  maxTokens: number;
  cacheEnabled: boolean;
  cacheTtlMs: number;
  retries: number;
  timeoutSeconds: number;
}

const configs: Record<Environment, CohereEnvConfig> = {
  development: {
    chatModel: 'command-r7b-12-2024',    // Fastest, cheapest
    embedModel: 'embed-v4.0',
    rerankModel: 'rerank-v3.5',
    maxTokens: 200,                       // Limit for dev
    cacheEnabled: false,                   // See real responses
    cacheTtlMs: 0,
    retries: 1,                            // Fail fast in dev
    timeoutSeconds: 30,
  },
  staging: {
    chatModel: 'command-r-08-2024',       // Mid-tier for testing
    embedModel: 'embed-v4.0',
    rerankModel: 'rerank-v3.5',
    maxTokens: 1000,
    cacheEnabled: true,
    cacheTtlMs: 5 * 60 * 1000,           // 5 minutes
    retries: 3,
    timeoutSeconds: 60,
  },
  production: {
    chatModel: 'command-a-03-2025',       // Best quality
    embedModel: 'embed-v4.0',
    rerankModel: 'rerank-v3.5',
    maxTokens: 4096,
    cacheEnabled: true,
    cacheTtlMs: 15 * 60 * 1000,          // 15 minutes
    retries: 5,
    timeoutSeconds: 120,
  },
};

function detectEnvironment(): Environment {
  const env = process.env.NODE_ENV ?? 'development';
  if (['development', 'staging', 'production'].includes(env)) {
    return env as Environment;
  }
  return 'development';
}

export function getCohereConfig(): CohereEnvConfig & { environment: Environment } {
  const env = detectEnvironment();
  return { ...configs[env], environment: env };
}

Step 2: Environment-Aware Client

// src/cohere/client.ts
import { CohereClientV2 } from 'cohere-ai';
import { getCohereConfig } from '../config/cohere';

let instance: CohereClientV2 | null = null;

export function getCohere(): CohereClientV2 {
  if (!instance) {
    const config = getCohereConfig();

    if (!process.env.CO_API_KEY) {
      throw new Error(`CO_API_KEY not set for ${config.environment} environment`);
    }

    instance = new CohereClientV2({
      token: process.env.CO_API_KEY,
      timeoutInSeconds: config.timeoutSeconds,
    });

    console.log(`[cohere] Initialized for ${config.environment} (model: ${config.chatModel})`);
  }
  return instance;
}

Step 3: Secret Management

# --- Local Development ---
# .env.local (git-ignored)
CO_API_KEY=trial-key-for-dev

# --- GitHub Actions (CI) ---
gh secret set CO_API_KEY --body "production-key-for-ci"

# --- AWS Secrets Manager ---
aws secretsmanager create-secret \
  --name cohere/staging/api-key \
  --secret-string "staging-production-key"

aws secretsmanager create-secret \
  --name cohere/production/api-key \
  --secret-string "prod-production-key"

# --- GCP Secret Manager ---
echo -n "staging-key" | gcloud secrets create cohere-api-key-staging --data-file=-
echo -n "prod-key" | gcloud secrets create cohere-api-key-prod --data-file=-

# --- HashiCorp Vault ---
vault kv put secret/cohere/staging api_key="staging-key"
vault kv put secret/cohere/production api_key="prod-key"

Step 4: Environment Guards

// Prevent expensive operations in development
function guardExpensiveOperation(operation: string): void {
  const config = getCohereConfig();

  if (config.environment === 'development') {
    // In dev, warn but don't block
    console.warn(`[cohere] ${operation} using trial key — limited to 20 calls/min`);
  }
}

// Prevent development keys in production
function validateKeyForEnv(): void {
  const config = getCohereConfig();
  const key = process.env.CO_API_KEY ?? '';

  if (config.environment === 'production' && key.length < 30) {
    throw new Error('Production requires a production API key (not trial)');
  }
}

Step 5: Per-Environment API Calls

import { getCohereConfig } from '../config/cohere';

export async function chat(message: string): Promise<string> {
  const config = getCohereConfig();
  const cohere = getCohere();

  const response = await cohere.chat({
    model: config.chatModel,     // Environment-specific model
    messages: [{ role: 'user', content: message }],
    maxTokens: config.maxTokens, // Environment-specific limit
  });

  return response.message?.content?.[0]?.text ?? '';
}

export async function embed(texts: string[]): Promise<number[][]> {
  const config = getCohereConfig();
  const cohere = getCohere();

  const response = await cohere.embed({
    model: config.embedModel,
    texts,
    inputType: 'search_document',
    embeddingTypes: config.environment === 'development' ? ['float'] : ['int8'], // Cheaper in prod
  });

  return response.embeddings.float ?? response.embeddings.int8;
}

Step 6: Docker Compose for Local Multi-Env Testing

# docker-compose.yml
services:
  app-dev:
    build: .
    environment:
      - NODE_ENV=development
      - CO_API_KEY=${CO_API_KEY_DEV}
    ports:
      - "3000:3000"

  app-staging:
    build: .
    environment:
      - NODE_ENV=staging
      - CO_API_KEY=${CO_API_KEY_STAGING}
    ports:
      - "3001:3000"

Output

  • Per-environment Cohere configuration (model, tokens, timeout)
  • Secret management across dev/staging/prod
  • Environment guards preventing misuse
  • Docker compose for local multi-env testing

Error Handling

IssueCauseSolution
Trial key in productionWrong secretValidate key length at startup
Rate limited in devTrial key limitsUse 20 calls/min budget
Model not foundTypo in configValidate model IDs at startup
Config merge failsMissing environmentDefault to development

Resources

Next Steps

For observability setup, see cohere-observability.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.