Picking an agent framework

The multi-agent framework landscape has fractured into tiers with distinct maturity profiles — not converged — and every major framework carries confirmed production failure modes that docs don't warn about.

Five categories of hook failures — silent non-firing, ignored decisions, platform breakage, data corruption, and architectural constraints — mean defense-in-depth across multiple events is required.

Streaming guardrails are architecturally broken by design -- OpenAI has marked the fix NOT_PLANNED.

LangGraph checkpoint serialization fails silently for non-primitive types — round-trips are lossy with no exception raised, across four documented modes since Jan 2026.