The Polyglot Protocol
A portable skill that gives coding agents a senior-engineer standard for writing code across multiple languages and frameworks, covering architecture, testing, security, performance, and validation.
The Polyglot Protocol is a portable skill for coding agents such as Codex, Claude Code, and OpenCode that defines engineering standards for polyglot codebases spanning several programming languages, frameworks, or runtimes. It is explicitly a protocol rather than a framework: it does not replace an agent's runtime, command system, planner, or orchestration layer, but instead defines the standards those systems should follow, and can act as a quality layer alongside frameworks such as the OAC Framework.
The stated problem is that coding agents write code quickly but without consistent engineering judgment, guessing at APIs, ignoring repository conventions, choosing the wrong language for a task, adding unnecessary dependencies, or skipping validation, especially as this compounds across multiple languages in one project. The protocol is meant to give an agent a repeatable way to inspect a repository, preserve its existing conventions, choose the right language, verify current tooling, and avoid inventing behavior, before it starts generating code.