Getting Started

Prerequisites

A coSPEC account (sign up for the beta)
An Anthropic API key (for the AI model)

1. Get Your API Key

In the dashboard, go to API Keys → Create API Key.

Keys use the format csk_live_* and are shown only once at creation. Store it securely.

See API Keys for details on key management.

2. Store Your Anthropic Key

Go to Integrations → Claude Code in the dashboard. Enter your Anthropic API key and click Connect.

coSPEC stores your key securely and injects it into the sandbox at runtime. The agent uses your Anthropic key to call Claude models directly.

3. Connect a Repository

Go to Integrations → GitHub in the dashboard and click Connect Account.

You'll be redirected to GitHub's authorization page. Grant access to the repositories you want coSPEC to work with.

See GitHub Integration for details.

4. Start Your First Run

Pick your tool and follow the guide:

Platform	Install	How Results Arrive	Best For
Zapier	Invite link	Automatic (step waits)	Simplest setup, one-step Zaps
Make	Invite link	Webhook trigger (separate scenario)	Multi-step scenarios
n8n	Community node	Polls automatically or webhook trigger	Self-hosted, AI agent workflows
REST API	None	Poll or webhook	Custom code, scripts
GitHub Actions	Coming soon	Workflow step	CI/CD pipelines
MCP	Coming soon	Tool call response	AI coding assistants

Quick test with curl

You can also create a run directly from the command line:

Terminal

curl -X POST https://api.cospec.io/v1/runs \
  -H "Authorization: Bearer csk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "repo": "your-org/your-repo",
    "prompt": "Add input validation to the signup form",
    "template": "your-template",
    "model": "sonnet"
  }'

Poll the run status until it completes:

Terminal

curl https://api.cospec.io/v1/runs/:id \
  -H "Authorization: Bearer csk_live_..."

Runs move through the lifecycle: pending → running → completed. When the run finishes, you'll see the output — text responses, branches, pull requests, or issues.

See the REST API guide for the full walkthrough.

What's Next

Learn more about Runs — lifecycle, outputs, and guardrails
Customize sandbox environments with Templates
Explore all endpoints in the API Reference