发布门禁
- must-include recall 不下降。
- 高意图 query 的空结果率不上升。
- 关键文档仍在预期 top 3 或 top 5。
- 任何有意的排序变化都有例子说明。
- 可回滚到旧索引、旧 embedding 模型或旧 ranker 配置。
Embedding drift monitoring runbook for RAG and agent search. Uses golden queries, recall@K, rank delta, and rollback gates.
MCP tool calling latency runbook for agents. Measures tools/list p95, separates server latency from network delay, and defines pause rules.
LLM prompt caching techniques for agents and apps. Covers stable prefixes, cache keys, TTLs, metrics, and cached-output validation.
Datadog LLM Observability traces OpenAI / Anthropic / Bedrock calls, tracks per-user cost, surfaces drift. Dashboards and span-level prompt view.
Expand-contract database migration checklist for agents. Covers additive schema changes, batched backfills, rollback, and contract gates.