Introducing Showboat and Rodney, so agents can demo what they’ve built

2026-02-10 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, medium

Summary

On February 10, 2026, two new CLI tools, Showboat and Rodney, were released to address the challenge of coding agents demonstrating their work. Showboat, a Go binary optionally wrapped in Python, helps agents construct Markdown documents that detail their code's functionality, automatically embedding command outputs and images. Rodney, built on the Go Rod library, provides CLI browser automation, enabling agents to control Chrome for web-based demonstrations, including opening URLs, executing JavaScript, clicking elements, and taking screenshots. These tools aim to reduce manual QA time by allowing agents to create verifiable demonstrations, minimizing the need for expensive QA agent swarms. The author developed both tools primarily using Claude Code for web via an iPhone app, emphasizing their utility in asynchronous coding agent environments.

Key takeaway

For Machine Learning Engineers integrating coding agents into development workflows, you should adopt tools like Showboat and Rodney to streamline agent output verification. These tools enable agents to autonomously generate clear, verifiable demonstrations of their code's functionality, significantly reducing the manual QA burden and ensuring delivered code actually works as intended. Incorporate these into your agent prompts to foster more reliable and transparent development cycles.

Key insights

Showboat and Rodney enable coding agents to autonomously demonstrate their work through verifiable documentation and browser automation.

Principles

• Code delivery requires proof of functionality.
• Automated demonstrations reduce manual QA.
• Agent tools should be self-documenting via help text.

Method

Agents use Showboat commands (init, note, exec, image) to build Markdown demos, with Rodney providing browser interaction (start, open, js, click, screenshot) for web-based features, all guided by comprehensive `--help` output.

In practice

• Use `showboat` to generate Markdown demos of agent-built features.
• Employ `rodney` for browser automation within agent workflows.
• Instruct agents to use `uvx tool --help` for self-guided tool usage.

Topics

• Coding Agents
• Software Demonstration
• Browser Automation
• CLI Tools
• Test-Driven Development

Code references

• simonw/showboat
• simonw/rodney
• simonw/go-to-wheel
• simonw/showboat-demos
• simonw/sqlite-history-json

Best for: Machine Learning Engineer, Software Engineer, AI Engineer, Prompt Engineer

Related on AIssential

• Harness engineering and agent feedback: Exploring AI coding sensors — Computational sensors provide deterministic feedback, enabling AI coding agents to self-correct and improve code quality autonomously.
• Amplitude ships 3x more production code with Cursor — Cloud-based autonomous agents significantly boost development velocity by enabling parallel execution and full environment access.
• Code smells for AI agents: Q&A with Eno Reyes of Factory — High-quality codebases are crucial for AI agent productivity, as agents excel at pattern recognition.
• Ollama on Steroids with OpenCode: A Headless Multi Agent Coding Workflow — A multi-agent system provides a reproducible, auditable, and private workflow for AI-driven code generation.
• What's next for coding agents? — AI agents are rapidly evolving with new interfaces, world-building capabilities, and autonomous communication platforms.
• How to Improve Claude Code Performance with Automated Testing — Automated and effective testing is crucial for maximizing coding agent efficiency and overcoming development bottlenecks.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.