Abstract
Existing multi-agent software development systems have proposed various forms of agent collaboration, including role-based collaboration and automated code reviews. However, many systems assume a common runtime, a central conversation server, or the same API family. Under these assumptions, LLM agents from different vendors cannot easily exchange messages directly from their own execution environments while dividing development and review work on a shared codebase.
This paper presents tap, a file-based collaboration protocol that allows Claude (Anthropic) and Codex (OpenAI) to collaborate on one codebase without shared memory or an identical runtime. The core of tap is a file-first design that preserves markdown files with metadata as original messages, combines a file inspection path (file communication, Tier 1) with real-time notification paths for Claude and Codex (real-time communication, Tier 2), and isolates work through separate git worktrees. Even if real-time notification fails or a receiver restarts, the message file remains available and the same content can be inspected again.
In a 27-day, 37-generation self-applied operation where tap was used to develop and review itself, we collected 209 tap-related pull requests and 717 operational artifacts. An analysis of 375 review artifacts showed that the share of reviews recording at least one defect or requested change was 69.8% for heterogeneous model pairs and 53.1% for homogeneous model pairs. These results demonstrate that tap, which combines file-based message preservation with real-time notification, operates effectively in a real production repository, and that combining heterogeneous models and execution environments can broaden review perspectives. tap is distributed as the open-source npm package @hua-labs/tap (v0.5.2).
Blogger's Review: The innovative design of the tap protocol makes collaboration between different LLMs more flexible and efficient, especially without shared memory, achieving high-quality code reviews. This approach not only enhances interaction among multiple models but also opens new possibilities for future LLM collaborations. Its open-source nature encourages community involvement and development, making it worthy of attention.