Autonomous pentesting with local LLMs, OPA policy enforcement, and deterministic tool execution. Zero data egress.
Clear separation of concerns. Intelligence reasons, execution runs tools. OPA sits between them—every task requires policy authorization.
Feed OpenAPI spec and business context. LLM builds an application model: roles, sensitive operations, control points.
LLM generates control validation hypotheses—not CVE lists. Focuses on authorization gaps, data isolation failures, workflow bypasses.
Link hypotheses into multi-step attack chains. Prioritize by business impact × feasibility.
Every task hits OPA before execution. Scope, time window, tool permissions, blocked parameters—all enforced at policy level.
package pentest.task
default allow = false
allow {
target_in_scope
within_time_window
tool_permitted
not blocked_by_roe
}
target_in_scope {
net.cidr_contains(
input.engagement.scope.cidrs[_],
input.task.target
)
}
within_time_window {
time.parse_rfc3339_ns(input.current_time) >=
time.parse_rfc3339_ns(input.engagement.window.start)
}
# Block dangerous flags
blocked_by_roe {
input.task.tool == "sqlmap"
input.task.params[_] == "--os-shell"
}Policies versioned and included in evidence bundles.
Customer role can approve loan applications via direct API call toPOST /api/loans/{id}/approvebypassing role-based controls.
/api/loans/{id}/approvesha256:7f3a91c2e8b4d5f6...We're working with banks and enterprises that need zero data egress and policy-provable security validation.