Skip to main content
Enterprise Integration

CSOAI API & SDK

Integrate AI safety governance directly into your systems. Programmatic access to compliance, council voting, and Watchdog alerts.

1. Get API Key

Generate your API key from the dashboard settings page.

2. Install SDK

Install our SDK for Python, JavaScript, or use REST directly.

pip install coai

3. Start Building

Register systems, run assessments, and integrate Watchdog alerts.

API Endpoints
Method
Endpoint
Description
POST
/v1/ai-systems
Register a new AI system for compliance tracking
GET
/v1/ai-systems
List all registered AI systems
GET
/v1/ai-systems/:id
Get details of a specific AI system
PUT
/v1/ai-systems/:id
Update an AI system
DELETE
/v1/ai-systems/:id
Delete an AI system
POST
/v1/compliance/assess
Run compliance assessment against frameworks
GET
/v1/compliance/frameworks
List supported compliance frameworks
GET
/v1/compliance/requirements
Get requirements for a framework
POST
/v1/council/submit
Submit to 33-Agent Council for review
GET
/v1/council/sessions
List council voting sessions
GET
/v1/council/sessions/:id
Get session details and votes
GET
/v1/watchdog/reports
Get public Watchdog reports
POST
/v1/watchdog/reports
Submit a Watchdog report
POST
/v1/webhooks
Configure webhook for real-time alerts
GET
/v1/webhooks
List configured webhooks
Code Examples
import coai

# Initialize the client
client = coai.Client(api_key="your_api_key")

# Register an AI system
system = client.ai_systems.create(
    name="My AI Chatbot",
    description="Customer service chatbot",
    system_type="chatbot",
    risk_level="limited"
)

# Run compliance assessment
assessment = client.compliance.assess(
    system_id=system.id,
    frameworks=["eu_ai_act", "nist_rmf"]
)

# Get compliance score
print(f"EU AI Act Score: {assessment.eu_ai_act.score}%")
print(f"NIST RMF Score: {assessment.nist_rmf.score}%")

# Submit to 33-Agent Council for review
session = client.council.submit(
    system_id=system.id,
    subject="Compliance Review Request",
    context=assessment.summary
)

# Check voting status
print(f"Council Decision: {session.decision}")
print(f"Votes: {session.votes}")
Webhooks & Real-time Alerts

Configure webhooks to receive real-time notifications when important events occur.

watchdog.report

Triggered when a new Watchdog report is submitted about your AI system

council.decision

Triggered when the 33-Agent Council reaches a decision on your submission

compliance.alert

Triggered when compliance status changes or new requirements are detected

API Pricing Tiers

Free

£0/mo

  • 1 AI system
  • 100 API calls/month
  • Basic compliance reports
  • Community support
Most Popular

Pro

£500/mo

  • 10 AI systems
  • 10,000 API calls/month
  • All frameworks (EU, NIST, TC260)
  • Webhooks & real-time alerts
  • Priority support

Enterprise

£5,000/mo

  • Unlimited AI systems
  • Unlimited API calls
  • Custom SDK integration
  • SSO/SAML authentication
  • 99.9% SLA guarantee
  • Dedicated support

Ready to integrate?

Get your API key and start building AI safety into your systems today.

Join the Byzantine Council

Help govern AI safety decisions

Security & Compliance

  • Training delivered through ISO 27001:2022 certified infrastructure
  • Powered by SOC 2 Type II compliant platform (CSGA)
  • Built on enterprise-grade security infrastructure

© 2026 CSOAI. All rights reserved.

CSOAI Legal Structure

CSOAI LTD (Council Safety of Artificial Intelligence LTD), Company Number: 16939677, registered at 86-90 Paul Street, London, EC2A 4NE, United Kingdom. As the world's first open-source FAA for AI, we operate as an independent Standards Body with 100% transparency. Our entire codebase, governance processes, and decision-making are publicly auditable. We maintain strict independence from all commercial AI vendors and technology companies. Our sole mission is to serve the public interest through independent AI safety oversight.

CEASAI Certification Standards

CEASAI (Certified Expert in AI Safety) is our professional certification program administered by CSOAI. CEASAI certification represents demonstrated competency in AI safety principles, compliance frameworks, and incident analysis. Certificates are valid for three years from issuance. CSOAI reserves the right to update certification requirements and may require recertification to maintain active status.

Independence & Transparency Statement

CSOAI is 100% independent with zero financial ties to OpenAI, Anthropic, Google, Microsoft, Meta, or any AI vendor. Our Byzantine Council uses 33 AI agents across 12 different providers to ensure no single entity controls safety decisions. All assessments are publicly auditable. All standards are openly developed. We answer only to the public interest—not shareholders, not AI companies, not governments. This is how safety should work.

Data & Privacy

Your data is protected under industry-standard security practices including encryption, secure authentication, and regular security audits. We comply with GDPR, CCPA, UK Data Protection Act 2018, and other privacy regulations. We never sell your personal information to third parties. For detailed information, please review our Privacy Policy, Cookie Policy, and Data Processing Agreement.

Cookie Preferences

We use cookies to enhance your experience

We use cookies and similar technologies to help personalize content, tailor and measure ads, and provide a better experience. By clicking "Accept All", you consent to the use of ALL cookies. You can manage your preferences by clicking "Customize".