Autonomous Agent

The Autonomous Security Agent is RedCloud’s AI-driven operator. It plans and executes security work within strict rules of engagement, choosing from a categorized tool matrix and running multiple parallel sessions, while the RedCloud Brain continuously learns from new intelligence.

In this section

• Overview — the six agent modules and how they fit together.
• Rules of Engagement — define scope, constraints, and objectives the agent must respect.
• Tool Matrix — the full set of tools, categorized by attack phase.
• Multi-Session — manage several parallel agent sessions.
• RedCloud Brain — the autonomous learning system with nightly/morning/evening cycles and CVE intelligence.

Governed autonomy

The agent never operates without guardrails. Rules of Engagement bound every session, the audit trail records every action, and tenant isolation is always enforced — autonomy is powerful, but never unsupervised.

Caution

Always review the Rules of Engagement before starting an agent session. Scope and constraints defined there are enforced for the entire session.