CBCL: Safe Self-Extending Agent Communication
TLDR
CBCL introduces a provably safe agent communication language that allows self-extending dialects while ensuring all messages remain within the DCFL class.
Key contributions
- CBCL constrains all agent messages, including runtime extensions, to the Deterministic Context-Free Language (DCFL) class.
- Agents can define, transmit, and adopt domain-specific "dialect" extensions as first-class, provably safe messages.
- Three machine-checked safety invariants prevent unbounded expansion, enforce resource limits, and preserve core vocabulary.
- Formalized in Lean 4 and implemented in Rust, demonstrating provably safe homoiconic protocol design.
Why it matters
This paper addresses the critical challenge of enabling extensible agent communication without sacrificing validation tractability. By formally bounding what agents can express, CBCL provides a foundational step towards safer, more auditable autonomous systems. This is vital for oversight as agents increasingly extend their own capabilities.
Original Abstract
Agent communication languages (ACLs) enable heterogeneous agents to share knowledge and coordinate across diverse domains. This diversity demands extensibility, but expressive extension mechanisms can push the input language beyond the complexity classes where full validation is tractable. We present CBCL (Common Business Communication Language), an agent communication language that constrains all messages, including runtime language extensions, to the deterministic context-free language (DCFL) class. CBCL allows agents to define, transmit, and adopt domain-specific "dialect" extensions as first-class messages; three safety invariants (R1--R3), machine-checked in Lean 4 and enforced in a Rust reference implementation, prevent unbounded expansion, applying declared resource limits, and preserving core vocabulary. We formalize the language and its safety properties in Lean 4, implement a reference parser and dialect engine in Rust with property-based and differential tests, and extract a verified parser binary. Our results demonstrate that homoiconic protocol design, where extension definitions share the same representation as ordinary messages, can be made provably safe. As autonomous agents increasingly extend their own communication capabilities, formally bounding what they can express to each other is a precondition for oversight.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.