Benjamin Zorn

Reshabh K Sharma, Shraddha Barke, Benjamin Zorn · uw

AI agents are increasingly embedded in real software systems, where they execute multi-step workflows through multi-turn dialogue, tool invocations, and intermediate decisions. These long execution histories, called agentic traces, make validation difficult. Outcome-only benchmarks can miss critical procedural failures, such as incorrect workflow routing, unsafe tool usage, or violations of prompt-specified rules. This paper presents AgentPex, an AI-powered tool designed to systematically evaluate agentic traces. AgentPex extracts behavioral rules from agent prompts and system instructions, then uses these specifications to automatically evaluate traces for compliance. We evaluate AgentPex on 424 traces from τ2-bench across models in telecom, retail, and airline customer service. Our results show that AgentPex distinguishes agent behavior across models and surfaces specification violations that are not captured by outcome-only scoring. It also provides fine-grained analysis by domain and metric, enabling developers to understand agent strengths and weaknesses at scale.

Reshabh K Sharma, Jonathan De Halleux, Shraddha Barke et al. · uw

Large language models (LLMs) are being used in many applications and prompts for these models are integrated into software applications as code-like artifacts. These prompts behave much like traditional software in that they take inputs, generate outputs, and perform some specific function. However, prompts differ from traditional code in many ways and require new approaches to ensure that they are robust. For example, unlike traditional software the output of a prompt depends on the AI model that interprets it. Also, while natural language prompts are easy to modify, the impact of updates is harder to predict. New approaches to testing, debugging, and modifying prompts with respect to the model running them are required. To address some of these issues, we developed PromptPex, an LLM-based tool to automatically generate and evaluate unit tests for a given prompt. PromptPex extracts input and output specifications from a prompt and uses them to generate diverse, targeted, and valid unit tests. These tests are instrumental in identifying regressions when a prompt is changed and also serve as a tool to understand how prompts are interpreted by different models. We use PromptPex to generate tests for eight benchmark prompts and evaluate the quality of the generated tests by seeing if they can cause each of four diverse models to produce invalid output. PromptPex consistently creates tests that result in more invalid model outputs than a carefully constructed baseline LLM-based test generator. Furthermore, by extracting concrete specifications from the input prompt, PromptPex allows prompt writers to clearly understand and test specific aspects of their prompts. The source code of PromptPex is available at https://github.com/microsoft/promptpex.

Shraddha Barke, Christian Poelitz, Carina Suzana Negreanu et al.

Large language models (LLMs) are rapidly replacing help forums like StackOverflow, and are especially helpful for non-professional programmers and end users. These users are often interested in data-centric tasks, such as spreadsheet manipulation and data wrangling, which are hard to solve if the intent is only communicated using a natural-language description, without including the data. But how do we decide how much data and which data to include in the prompt? This paper makes two contributions towards answering this question. First, we create a dataset of real-world NL-to-code tasks manipulating tabular data, mined from StackOverflow posts. Second, we introduce a cluster-then-select prompting technique, which adds the most representative rows from the input data to the LLM prompt. Our experiments show that LLM performance is indeed sensitive to the amount of data passed in the prompt, and that for tasks with a lot of syntactic variation in the input table, our cluster-then-select technique outperforms a random selection baseline.

Daniel W. Barowy, Emery D. Berger, Benjamin Zorn

Spreadsheets are one of the most widely used programming environments, and are widely deployed in domains like finance where errors can have catastrophic consequences. We present a static analysis specifically designed to find spreadsheet formula errors. Our analysis directly leverages the rectangular character of spreadsheets. It uses an information-theoretic approach to identify formulas that are especially surprising disruptions to nearby rectangular regions. We present ExceLint, an implementation of our static analysis for Microsoft Excel. We demonstrate that ExceLint is fast and effective: across a corpus of 70 spreadsheets, ExceLint takes a median of 5 seconds per spreadsheet, and it significantly outperforms the state of the art analysis.