Consider AIUC standard (AIUC-1)

Link - https://aiuc.com/

AIUC-1, the first dedicated security, safety, and reliability standard for AI agents, was officially introduced in 2025 by the Artificial Intelligence Underwriting Company (AIUC) as a compliance framework for AI agent vendors. The standard was designed to provide enterprises with confidence in deploying agentic AI systems by implementing over 50 technical, operational, and legal safeguards across six domains: Safety, Security, Reliability, Accountability, Data & Privacy, and Society.
The launch aimed to address the growing risks associated with AI agents, including hallucinations, harmful outputs, prompt injection, and unauthorized actions, which were not fully covered by existing certifications like SOC 2 or ISO 42001.
AIUC-1 operationalizes high-level frameworks such as NIST AI RMF, MITRE ATLAS, and OWASP Top 10 for agentic applications into concrete, testable controls.
Since its launch, AIUC-1 has been updated quarterly to keep pace with evolving AI capabilities, enterprise adoption trends, and regulatory requirements, ensuring that the standard remains current and relevant for organizations deploying AI agents.
Certification involves rigorous technical testing, ongoing retesting, and annual audits to validate compliance and safety, making it a SOC-like assurance framework specifically for AI agents.

Key Points
Launched: 2025
Organisation: Artificial Intelligence Underwriting Company (AIUC)
Purpose: Security, safety, and reliability standard for AI agents
Controls: 50+ technical, operational, and legal safeguards
Domains Covered:
Safety
Security
Reliability
Accountability
Data & Privacy
Society
Addresses Risks:
Hallucinations
Harmful outputs
Prompt injection
Unauthorized actions
References Frameworks:
NIST AI RMF
MITRE ATLAS
OWASP Top 10 for Agentic Applications
Assurance Model:
Technical testing
Ongoing retesting
Annual audits
SOC-like certification approach for AI agents

Start by reviewing the linked AIUC-1 standard and comparing its six domains and controls with this repository’s existing MCP governance framework. Determine where the standard should be represented and define an explicit acceptance scope for adding or referencing it; the issue currently names no files, tests, or completed output.