AI 智能体 model-serving-security 更新于 2026-10-27

Model Serving · 安全规则

Security rules for Model Serving: boundary validation, server-side authorisation and zero secret leakage.

✓ 应当这么做

✓ Validate and normalise every input at the boundary it enters the system
✓ Authorise on the server against the authenticated principal, never on client claims
✓ Keep secrets out of repositories, logs and error messages

42 篇

Vector search rules: chunking strategy, recall evaluation and index rebuild path

Feature store rules: point-in-time correctness and train/serve parity

Skill design rules: narrow triggers, explicit boundaries and verifiable outputs

Tool-call rules: server-side authorisation and deny-by-default allow-lists

Memory rules: scoped retrieval, expiry and user-visible deletion

RAG rules: citation-grade retrieval and evaluated recall before shipping