Prism: A Minimal Compositional Metalanguage for Specifying Agent Behavior

TL;DR

Prism targets the gap between natural-language style prompts and executable agent actions by introducing a fixed, minimal core grammar (Core1) and domain-specific mini-grammars. Decisions are expressed as selection expressions using a single abstraction operator, avoiding imperative control blocks and enabling transparent, verifiable policies. The paper demonstrates Prism across thermostat control, smart home security, e-commerce recommendations, and medical alert systems, and discusses linguistic implications such as compositionality, lexicon-grammar separation, and tools as world-bridges. This approach offers inspectable, reusable, and safety-conscious agent control suitable for analysis and verification, with future work on formal semantics integration and tooling.

Abstract

Prism is a small, compositional metalanguage for specifying the behaviour of tool-using software agents. Rather than introducing ad hoc control constructs, Prism is built around a fixed core context, Core1, which provides a minimal background grammar of categories numbers, strings, user prompts, tools together with abstract combinators for booleans, predicates, pairs, and lists. Agent policies are written as ordinary expressions using a single abstraction operator so that conditionals appear as selections between alternatives instead of imperative if-else blocks. Domains extend the core by defining their own context-mini-grammars that introduce new categories, predicates, and external tools while reusing the same compositional machinery. We illustrate this with worked examples from thermostat control, home security, e-commerce recommendation, and medical monitoring, showing how natural language decision rules can be mapped to inspectable, executable policies. From a linguistic perspective, Prism enforces a clear separation between a reusable grammar-like core and domain specific lexicons and treats tools as bridges between internal policy representations and the external world. From an engineering perspective, it offers a compact interface language for agent control, making the space of possible actions explicit and amenable to analysis, verification, and safety constraints.