For Developers

The Developer Toolkit

SDK & Agent API

Every agent on CEO.ai has its own unique API key. Create a client instance, send prompts, and get RAG-powered responses. Zero external dependencies — uses only Node.js built-in modules.

sdk-usage.js

const { CeoAI } = require('@ceo-ai/sdk');  
  
const ceo = new CeoAI({  
  apiKey: 'sk_live_your_api_key_here',  
  // endpoint, timeout, pollInterval, pollTimeout — all optional  
});  
  
// Send a prompt and wait for the result  
const { response, metadata } = await ceo.promptAndWait(  
  'What was our Q4 revenue?'  
);  
  
console.log(response);  
// => { answer: "Q4 revenue was $12.5M, representing a 15% increase..." }  
  
console.log(metadata);  
// => { agentId: '...', agentName: 'Financial Analyst', model: 'claude-sonnet-4-5', estimatedCredits: 10 }

conversation.js

// First message  
const first = await ceo.promptAndWait('What was our Q4 revenue?');  
  
// Follow-up with context  
const second = await ceo.promptAndWait('How does that compare to Q3?', {  
  conversationHistory: [  
    { role: 'user', content: 'What was our Q4 revenue?' },  
    { role: 'assistant', content: JSON.stringify(first.response) }  
  ]  
});

Full SDK Reference

CLI — `@ceo-ai/cli`

Send prompts, manage conversations, add RAG knowledge, and configure everything from your terminal.

terminal

# Install globally  
npm install -g @ceo-ai/cli  
  
# Configure your API key  
ceo configure --key sk_live_your_api_key_here  
  
# Send a prompt and wait for the result  
ceo prompt "What was our Q4 revenue?" --poll  
  
# Start a conversation (persists to a local JSON file)  
ceo chat "Analyze our marketing spend" -c marketing.json  
  
# Continue the conversation with full history context  
ceo chat "Break it down by channel" -c marketing.json  
  
# List all conversations  
ceo conversations

Key `ceo prompt` flags

Flag	Description
`--poll`	Wait for results instead of returning presigned URL
`--no-rag`	Disable RAG mode (RAG enabled by default)
`-o, --output <file>`	Write results to a file
`--json`	Raw JSON output (useful for piping)
`-k, --key <apiKey>`	Override configured API key for a single request
`--poll-interval <ms>`	Polling frequency (default: 2000ms)
`--poll-timeout <ms>`	Max wait time (default: 120000ms)

Custom command name

Personalize the CLI command to anything you want:

ceo configure --name jules  
ceo alias:setup  
# Follow the instructions, then:  
jules chat "What's the status of our project?"

Full CLI Reference

RAG Training via CLI

Forget clunky web interfaces. The CEO.ai CLI lets you manage RAG knowledge from your terminal. Add single files or ingest entire directories.

terminal — RAG management

# Add a single file to an agent's knowledge  
ceo addRag ./specs/api-patterns.md  
  
# Add an entire directory of files  
ceo addRagDirectory ./documentation/

Supports: .txt, .md, .pdf, .json, .yaml, .py, .js, .ts, and more. The CLI handles parsing and indexing.

CLI Reference

SDK & npm Package

npm install @ceo-ai/sdk

The SDK wraps every API endpoint with typed methods and handles authentication, error handling, and retry logic. Zero external dependencies. TypeScript support included. ESM and CJS exports. Node.js 16+.

npmjs.com/package/@ceo-ai/sdk SDK Reference

Build With It

AI-powered products

Train an agent on domain knowledge, give it an API key, embed it in your product. AI in your SaaS without ML infrastructure.

Messaging bots

Build agents that power WhatsApp, Telegram, or Slack bots — with real business intelligence via RAG, not generic chatbot responses.

Internal tools

Use the CEO Agent to one-shot full-stack internal apps. Frontend, backend, database, deployment configs. Committed to GitHub.

Integration layers

Chain agents into workflows that connect your existing platforms via API. Custom integration logic powered by AI.

AI agent businesses

Create specialized AI agents, expose them via API, sell access. Or build client-facing AI solutions. The Agent API is your infrastructure.

Express.js / backend integration

The SDK integrates cleanly with Express, Fastify, or any Node.js server. Async or sync endpoints, typed error handling.

Your CEO Found This. Here's Why You Should Be Excited.

If you're the developer (or the technical co-founder) on a team where the CEO is driving AI adoption — here's what CEO.ai means for you:

You're not losing control.

Everything commits to GitHub. You can see every file, every commit, every agent's contribution. The CEO Agent doesn't replace your judgment — it gives you a massive head start.

You get proper developer tools.

CLI, SDK, typed API, npm package. This isn't a no-code tool pretending to be developer-friendly. The programmatic access is real and comprehensive.

RAG training is actually fast.

Ingest your entire codebase documentation, your API specs, your infrastructure patterns — via ceo addRagDirectory ./docs/. Now the architect agents actually know your stack.

The CEO Agent saves you from grunt work.

Boilerplate. Migrations. Terraform configs. CRUD endpoints. Lambda functions. Let the agents handle the boring 80%. You focus on the interesting 20%.

You can rate and improve the output.

This isn't a black box. You rate each agent's work, and the system learns. Over time, the agents produce code that increasingly matches your standards and preferences.

Resources

Full API Documentation	docs.ceo.ai
CLI Reference	/developers/cli
SDK Reference	/developers/sdk
npm Package	@ceo-ai/sdk
Quickstart Guide	Getting Started in 5 Minutes
API Status	status.ceo.ai

Start Building

$ npm install @ceo-ai/sdk

$ npm install -g @ceo-ai/cli

View Documentation Create Your Account

Every plan includes Agent API access. CLI and SDK available on all plans. CEO Agent API available on Enterprise.

CLI, SDK, and API Access to a Full Agent Orchestration Platform