Design and review Adyen Checkout, merchant references, idempotency, HMAC webhooks, asynchronous results, modifications, and tests.
Design trajectory-level agent security evals for Goal compliance, contract boundaries, goal hacking, unsafe tools, and long-horizon behavior.
Review agent sandboxes, transitive egress, privilege escalation, lateral movement, shared infrastructure, monitoring, and kill switches.
Compare declared agent contracts with independent runtime enforcement across tools, files, processes, networks, secrets, approvals, and limits.
Threat-model agents as untrusted principals across prompts, tools, MCP, memory, delegation, runtime infrastructure, and external systems.
Review AI-assisted changes before execution for automation boundaries, human approval, affected-system criticality, and audit evidence.
Assess correlation of logs, metrics, traces, events, and incidents to reduce noise, improve root-cause analysis, and lower alert fatigue.
Review alerts for actionability, clear symptoms, runbook links, severity, ownership, SLO relation, deduplication, escalation, and remediation suitability.