🤖 AI Summary
This work addresses the lack of effective control over privilege escalation in existing intelligent systems during dynamic code generation and execution. We propose a "controlled metaprogramming" paradigm that treats program representations as first-class values, decoupling code generation from execution through purely syntactic manipulation and structural inspection mechanisms. Specifically, we reconceptualize eval—not as a language primitive—but as a controlled effect subject to validation against policies, capabilities, and resource constraints. Leveraging formal methods, we define pure syntactic evaluation and controlled materialization judgments, implementing them in MashinTalk, a domain-specific language compiled to BEAM bytecode. Our theoretical guarantees—encompassing purity of syntactic operations, non-bypassability, and boundary preservation—are formally verified within a suite of 454 machine-checked theorems in Rocq.
📝 Abstract
AI systems increasingly synthesize executable structure at runtime: LLMs generate programs, agents construct workflows,self-improving systems modify their own behavior. In classical homoiconic and staged languages, the transition from coderepresentation to execution is unrestricted. eval is a language primitive, not a governed operation. We argue that ingovernedintelligent systems, this transition is an authority amplification: it converts symbolic structure into executableauthority andmust be mediated like any other effect. We present governed metaprogramming, a language design where programrepresentations(machine forms) are first-class values, form manipulation is pure computation, and materialization (the transition fromform toexecutable machine) is a governed effect subject to structural inspection. The governance system analyzes the proposedprogram'scapability requirements, policy compliance, and resource estimates before permitting execution. We formalize twojudgments: pureform evaluation (which emits no directives) and governed materialization (which emits exactly one governed directive). Weprovethree properties: purity of form manipulation, the no-bypass theorem, and boundary preservation. We implement the designinMashinTalk, a DSL for AI workflows compiling to BEAM bytecode, and report on integration with 454 existingmachine-checked Rocqtheorems. The central contribution is reclassifying eval from a language primitive into a governed effect.