EDATracer: An Agentic Framework for Large-Scale EDA Artifact Analysis

2026-08-02

AI Summary

This work addresses the challenge in chip design where heterogeneous artifacts generated by EDA tools—such as source files, logs, and netlists—are fragmented and difficult to correlate, significantly impeding debugging and optimization. The paper proposes the first evidence-driven analysis framework tailored for large-scale EDA outputs, integrating knowledge graphs with semantic vector indexing to establish a unified representation that enables large language model agents to perform cross-artifact retrieval and reasoning. The contributions include the release of the first public benchmark for cross-artifact analysis and an open-source chip design dataset totaling 18.9 GB. On a 90-question benchmark, the proposed method outperforms Cursor and Claude Code by 6.4 and 7.2 percentage points in average pass@1 accuracy, respectively, while reducing token consumption by 2.0–3.2×.

Abstract

Modern chip design relies on electronic design automation (EDA) tools that generate large, heterogeneous artifacts, including source files, scripts, logs, netlists, and reports. Analyzing these artifacts is critical for debugging, optimization, and design-flow understanding, but remains difficult because relevant evidence is often distributed across many artifact types and design stages. Although LLM agents show promise for EDA assistance, existing approaches lack public benchmarks for large-scale cross-artifact analysis and often struggle to ground reasoning in tool-generated evidence. We present EDATracer, an agentic framework for evidence-grounded EDA artifact analysis. EDATracer organizes design artifacts into a knowledge graph paired with a semantic vector index, enabling LLM agents to retrieve evidence across source files, logs, netlists, and reports. We curate an 18.9 GB dataset of 2,787 synthesizable open-source chip designs and introduce a 90-question benchmark spanning factual, statistical, and reasoning tasks. Across evaluated agents, EDATracer achieves the best pass@1 accuracy, outperforming Cursor and Claude Code by 6.4 and 7.2 percentage points on average, while using 2.0-3.2x fewer tokens.

Problem

• EDA artifact analysis
• cross-artifact reasoning
• evidence grounding
• chip design debugging
• large-scale benchmark

Innovation

• agentic framework
• knowledge graph
• semantic vector index
• EDA artifact analysis
• large-scale benchmark