CodeMem: Architecting Reproducible Agents via Dynamic MCP and Procedural Memory

TL;DR

CodeMem introduces an architecture that leverages procedural memory to enhance the reliability and reproducibility of AI agents performing repetitive tasks, addressing instability issues inherent in probabilistic models.

Contribution

The paper presents CodeMem, a novel architecture that uses procedural memory to enable deterministic and reproducible agentic workflows built on large language models.

Findings

Improves agent reliability and reproducibility.

Reduces probabilistic instability in task execution.

Enables reusable and deterministic workflows.

Abstract

Current tool-using AI agents suffer from limited action space, context inefficiency, and probabilistic instability that makes them unsuitable for handling repetitive tasks which are otherwise reliably and efficiently tackled by agentic workflows built on platforms like n8n and Zapier. Earlier works like CodeAct, DynaSaur, Code Mode have tried to tackle the first two issues by using the whole Python language as its action space: The number of tools that the agent can call becomes infinite. Python code blocks can execute complex actions into a single step and print only relevant results which helps in keeping the context lean. However, the probabilistic instability issue still remains, as for the same task in the same environment, the agent can follow different trajectories due to the probabilistic nature of LLMs. Therefore, we need procedural memory for consistency and reliability. This paper proposes CodeMem, an architecture to implement procedural memory via code which can be used to build and run reusable agentic workflows with deterministic reliability.