Three Extension Injection Points

Description

From the design guide: every agent loop has three places where extensions can intervene. (1) assemble() — what the model sees (instructions, tool schemas). (2) model() — what the model can reach (available tools). (3) execute() — whether/how an action runs (permission gates, pre/post hooks). A useful cross-framework lens for classifying extension mechanisms.

Key claims

Relations

Sources

src-20260423-fd74bb3399e6