Documentation

Build with CogStorm API

Everything you need to integrate CogStorm into your workflows

Getting Started

Install the SDK and deploy your first agent in under 5 minutes.

Installation

bash

npm install @cogstorm/sdk

Quick Start

typescript

import { CogStorm } from '@cogstorm/sdk';

const client = new CogStorm({
  apiKey: process.env.COGSTORM_API_KEY,
});

// Deploy an agent
const agent = await client.agents.deploy({
  type: 'finance-analyst',
  name: 'Invoice Processor',
  config: {
    model: 'cogstorm-v3',
    memory: true,
  },
});

// Assign a task
const task = await client.tasks.create({
  agentId: agent.id,
  type: 'process-invoice',
  input: { documentUrl: 'https://...' },
});

Authentication

All API requests require a Bearer token. Include your API key in the Authorization header.

bash

# Include your API key in every request
curl https://api.cogstorm.com/v1/agents \
  -H "Authorization: Bearer cs_live_your_api_key" \
  -H "Content-Type: application/json"

Live Keys

Prefix: cs_live_ — Use in production. Requests are billed.

Test Keys

Prefix: cs_test_ — Use in development. No charges.

Security tip: Never expose API keys in client-side code. Use environment variables and server-side requests only.

Agents API

Create, manage, and monitor AI agents programmatically.

Tasks API

Assign work to agents and track task execution in real time.

Workflows API

Chain agents together into multi-step workflows with conditional logic.

Webhooks

Receive real-time notifications when agents complete tasks, encounter errors, or change status.

Signature verification: All webhook payloads include an X-CogStorm-Signature header. Always verify signatures before processing events.

SDKs

Official client libraries for your language of choice.

JavaScript / TypeScript

v3.2.1

npm install @cogstorm/sdk

Python

v3.1.0

pip install cogstorm

Go

v2.8.0

go get github.com/cogstorm/cogstorm-go

Ruby

v2.5.2

gem install cogstorm

Rate Limits

Rate limits are applied per API key. Upgrade your plan for higher throughput.

Tier	Requests / min	Agents	Concurrent Tasks
Free	60	3	5
Pro	600	25	50
Scale	6,000	200	500
Enterprise	Unlimited	Unlimited	Custom

Exceeding rate limits returns a 429 Too Many Requests response with a Retry-After header.