Ujex

Engineering · Ujex

Ujex trust engineering

Deep dives into agent passports, scoped mandates, approval gates, signed evidence, and production integration.

  1. Scoped credentials for AI agents: separating identity from authority

    Why a valid agent identity is only the start, and how expiring mandates constrain action, resource, budget, and consent.Read article →
  2. Benchmarking an agent trust layer: denial latency, replay safety, and evidence cost

    A reproducible benchmark plan for the controls that sit between an agent request and a real side effect.Read article →
  3. Integrating the Ujex MCP server in production

    From local emulators to agent enrolment, mandate checks, approval-gated tools, signed callbacks, and offline evidence verification.Read article →