KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #ai-agent #developer-tools #codex #typescript #cli #deepseek-harness #claude-code #ai-agents #mcp #rust

LEGO-An-LLM-Skill-Based-Front-End-Design-Generation-Platform

★ 29 Open GitHub ↗

ISEDA'26 LEGO official Repo

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# LEGO-An-LLM-Skill-Based-Front-End-Design-Generation-Platform
Existing LLM-based EDA agents are often isolated task-specific systems, leading to repeated engineering effort and limited reuse of successful design and debugging strategies. We present LEGO, a unified skill-based platform for front-end design generation that decomposes the digital front-end flow into six independent steps and represents every agent capability as a standardized, composable circuit skill within a plug-and-play architecture. To construct this skill library, we survey more than 100 papers, select 11 representative open-source projects, and extract 42 executable circuit skills organized into a six-step finite state machine formalization. Circuit Skill Builder automates skill extraction with linear scalability, while Agent Skill RAG achieves sub-millisecond retrieval without relying on embedding models. Empirical evaluation on a hard subset of 41 VerilogEval v2 problems that gpt-5.2-codex (xhigh) fails to solve demonstrates that individual circuit skills constructed within LEGO raise Pass@1 from 0.000 to 0.805, achieving an 80.5% gain over the baseline. Cross-project skill compositions also reach 0.805 Pass@1, outperforming hierarchy-verilog by 14.6%, VerilogCoder by 2.5%, and matching MAGE, demonstrating that modular skill composition enables both effective and flexible RTL design automation.

ISEDA'26 LEGO official Repo

The Skills are coming soon.

Contact: jinchenglou@stu.pku.edu.cn

LEGO-An-LLM-Skill-Based-Front-End-Design-Generation-Platform

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README