MCP 已把 Agent 身份列为协议重点。本文区分现行授权、企业扩展、工作负载身份、委托与 DPoP 的真实成熟度。
MCP agent identity is becoming a protocol priority. This guide separates stable authorization from draft workload identity, delegation, and DPoP work.
产品的 AI 可用性要看 Agent 能否完成真实任务,并通过外部状态、恢复能力、成本与回归稳定性验证。
Test whether AI agents can use your product with dynamic task evals that measure correct state changes, recovery, cost, and regression stability.
Flint 在 AI 意图和图表代码之间增加语义中间层,使可视化生成可编辑、可迁移、可测试。
Flint adds a semantic intermediate representation between AI intent and chart code, making visualization generation editable, portable, and testable.
HANDBOOK.md 说明规则进入 Context 仍不足以形成控制。本文给出规则检索、执行前判定、提交门控与事后核验四层方案。
HANDBOOK.md shows why policy in context is not control. Build AI agent compliance with retrieval, preflight decisions, commit gates, and verification.
MCP Elicitation 把工具执行中的人工决策变成可恢复的协议状态。本文讲清 form、URL、安全边界、状态持久化与重试设计。
MCP Elicitation turns missing input into recoverable control flow. Learn form vs URL mode, state handling, security boundaries, and retries.