Secure AI Atlas SECURITY & GOVERNANCE

EN / ES

AI Coding Agent

Mapped through AI Capability → Exposure Surface → Control Layer → Governance Layer.

AI Capability

Code generation
Repository access
Shell execution
Dependency installation

Exposure Surface

Source code
Secrets
CI/CD
Package registries

Risk

prompt injection excessive agency agentic supply chain compromise insecure tool invocation

Control

off agent authorization for tool calls human approval high impact input output validation agents prompt output logging

Framework

owasp agentic top 10not verified
owasp llm top 10 2026not verified
mitre atlasnot verified
nist ai rmfnot verified

Open questions

OQ-MCP-GOVERNANCE
OQ-AGENT-IDENTITY

This mapping is deterministic: it comes from relationships already present in the ATLAS catalogue. No model was used to infer it.