JSON Strategy Templates vs Executable WASM Genes: Two Paths for AI Agent Evolution

A pattern is emerging across AI agent infrastructure: modular capabilities that agents can discover, install, and share . Different projects call these units different things — skills, tools, capsules, genes — but the idea is converging. Agents shouldn't be monolithic. They should assemble capabilities from a shared ecosystem. Where projects diverge — sharply — is on a question that looks minor but determines everything downstream: what is the capability unit made of? One answer: a JSON document. A structured strategy template that an LLM reads, interprets, and acts on. Another answer: a compiled WASM binary. An executable program that runs in a sandbox with deterministic inputs and outputs. This isn't a taste preference. It's an architectural fork that determines what "evolution" can actually mean for AI agents. The JSON Template Approach The JSON strategy model works like this: a capability is encoded as a structured document containing a problem description, trigger conditions, a re