Composition without a Host: Carrying the Spatiotemporal-Composability Guarantee of Plugin-Based Agents across Processes
Not on the map. Why?
Abstract (arXiv)
Plugin-based agents assemble capabilities at runtime, and the spatiotemporal-composability calculus proves a reversibility guarantee for this assembly. However, the guarantee is carried by a single process, which confines all components, sessions, and recovery records to one failure domain, where a fault spreads past the plugin boundary, and process death interrupts every session the process hosts. Resting only on the hypotheses the calculus already states and the stateless interface of the model call, this paper relaxes the single-process restriction of the calculus to an arbitrary assignment of components and records to processes, gives four sufficient conditions, and proves with Theorem 1, derived from the four lemmas, that the reversibility guarantee holds across processes when these conditions are met. Based on Theorem 1, this paper constructs Logos, a cross-process plugin-based agent in the peer-process and name-routed form of ROS, where a plugin is a process, the router holds only a rebuildable routing table, and the session state needed for recovery lives in an append-only transcript owned by no process. Under one fault on two hundred benchmark tasks across three configurations, the single-process reference lost every session and scored 1.5 percent on the official validator, the MCP configuration kept its sessions while spending 1099 calls on a dead endpoint, and Logos kept every session alive, wasted zero calls, and succeeded on 120 tasks against 102 for both configurations combined. At the mechanism level, eighty sessions terminated at four points of the tool-call cycle all resumed with no repeated action, 3,500 concurrent calls paired with zero violations, and one bus hop cost 1 in 823 of the model's first token. The results show that the reversibility guarantee holds across processes and that assembly itself can leave the host process.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 8 of its 34 references that are in the map. First posted on arXiv 28 Aug 2026.
Builds on
8 of 34 references.
References · 8
- 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
- 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
- 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,332
- 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agentscited by 1,002
- 2025Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directionscited by 265
- 2025A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)cited by 101
- 2024AIOS: LLM Agent Operating Systemcited by 97
- 2024GoEX: Perspectives and Designs Towards a Runtime for Autonomous LLM Applicationscited by 23
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Revisable by Design: A Theory of Streaming LLM Agent Execution
- 2026Revocation Is Not Complete at Receipt: An Executable Outcome Contract for Distributed Agent-Tool Gateways
- 2026LogAct: Enabling Agentic Reliability via Shared Logs
- 2026DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving
- 2026Agent Mesh: Reliability Primitives for Non-Idempotent Agent Delegation - Identity Adequacy and Evidence Adequacy
- 2026The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World
- 2026DART: Semantic Recoverability for Structured Tool Agents
- 2026Harness Engineering as Categorical Architecture