AgentXRay: White-Boxing Agentic Systems via Workflow Reconstruction

TL;DR

AgentXRay introduces a search-based method to reconstruct interpretable workflows from black-box agentic systems, enhancing transparency and control without requiring internal model access.

Contribution

It formulates workflow reconstruction as a combinatorial optimization problem and employs Monte Carlo Tree Search with pruning to produce editable white-box workflows.

Findings

AgentXRay achieves higher proxy similarity than unpruned search.

It reduces token consumption during workflow reconstruction.

The method enables deeper exploration within fixed iteration budgets.

Abstract

Large Language Models have shown strong capabilities in complex problem solving, yet many agentic systems remain difficult to interpret and control due to opaque internal workflows. While some frameworks offer explicit architectures for collaboration, many deployed agentic systems operate as black boxes to users. We address this by introducing Agentic Workflow Reconstruction (AWR), a new task aiming to synthesize an explicit, interpretable stand-in workflow that approximates a black-box system using only input-output access. We propose AgentXRay, a search-based framework that formulates AWR as a combinatorial optimization problem over discrete agent roles and tool invocations in a chain-structured workflow space. Unlike model distillation, AgentXRay produces editable white-box workflows that match target outputs under an observable, output-based proxy metric, without accessing model parameters. To navigate the vast search space, AgentXRay employs Monte Carlo Tree Search enhanced by a scoring-based Red-Black Pruning mechanism, which dynamically integrates proxy quality with search depth. Experiments across diverse domains demonstrate that AgentXRay achieves higher proxy similarity and reduces token consumption compared to unpruned search, enabling deeper workflow exploration under fixed iteration budgets.