Arcadia-1/virtuoso-bridge-lite

Bridge between LLM-Agent and Cadence Virtuoso. A new infrastructure for Agentic Analog and Mixed-Signal Design.

What it solves

Virtuoso-bridge-lite provides a bridge between modern AI agents (like Claude Code or Cursor) and Cadence Virtuoso, the industry-standard tool for analog and mixed-signal IC design. It automates the tedious manual process of schematic editing, layout generation, and simulation setup, allowing LLM-driven agents to programmatically control the design flow.

How it works

The project implements a TCP-based SKILL client that communicates with a daemon running inside Virtuoso. It uses ipcBeginProcess and evalstring to execute SKILL commands. To handle remote environments, it manages SSH tunnels and jump hosts, ensuring a secure and stable connection between the local AI agent and the remote EDA server. It also includes a built-in Spectre simulator runner and a PSF (Parameter Storage Format) ASCII parser to bring simulation results back into Python for analysis.

Who it’s for

  • Analog/Mixed-Signal Designers: Those looking to automate repetitive design tasks using Python or AI agents.
  • AI Agent Developers: Developers building tools for automated chip design who need a reliable way to interface with Cadence tools.
  • EDA Tooling Engineers: Engineers needing a scalable, multi-server infrastructure for distributed design clusters.

Highlights

  • Deep Virtuoso Integration: Full control over Schematics, Layout, Maestro, and Spectre simulations.
  • AI-Native Design: Ships with pre-defined skill files for agents (Claude Code, Cursor) to immediately understand how to operate the bridge.
  • C-first Lifecycle: A comprehensive CLI for managing tunnels, checking daemon health, and executing SKILL expressions directly from the shell.
  • Scalable Architecture: Supports multi-server, multi-session setups with flexible SSH routing (OpenSSH or Paramiko).
  • Automated Snapshots: Ability to dump focused Maestro sessions, including setup and simulation artifacts, to local folders for agent review.

Related

  • Project
    stripe/ai

    A collection of SDKs and plugins that integrate Stripe's billing and payment infrastructure with LLMs and AI agent frameworks.

  • Project
    kirodotdev/Kiro

    An agentic AI development platform that helps developers plan, build, and review software by turning prompts into executable specs and using parallel agents across multiple interfaces.

  • Project
    yaojingang/yao-meta-skill

    A governance and engineering system that transforms rough workflows and prompts into reusable, cross-platform AI agent skill packages with built-in evaluation and release gates.

  • Project
    MayDay-wpf/snow-cli

    An agentic coding assistant for the terminal that integrates AI-driven development, command execution, and multi-agent collaboration directly into the CLI.

  • Dispatch
    vLLM Semantic Router: Enhancing Model Performance via Micro-Agent Collaboration

    vLLM introduces the Semantic Router, a serving-layer primitive that transforms a single model API call into a bounded collaboration of micro-agents to outperform frontier models on complex benchmarks.