Poimandres Plugin Gives Opus 5.5 189 Scenes

// 1h agoOPENSOURCE RELEASE

Poimandres Plugin Gives Opus 5.5 189 Scenes

The open-source plugin connects Claude Code to pmndrs’ MCP server, exposing React Three Fiber, Drei, and Zustand documentation plus working example sources. Opus 5.5 can start from real demos, clone their code, and restyle them instead of generating another gray-box scene.

// ANALYSIS

This is a retrieval and workflow upgrade more than a new 3D-generation breakthrough: curated examples give agents a reliable starting point and reduce hallucinated APIs.

–The MCP server returns demo source, live URLs, asset attribution, and dependency versions.
–The plugin teaches Claude to consult working examples before writing scenes from scratch.
–MIT licensing and GitHub installation make the workflow accessible to developers.
–The reported 189-scene count is powerful, though pmndrs’ current documentation lists 167 demos.
–Agencies still own art direction, performance tuning, responsive behavior, and production QA.

// TAGS

poimandres-claude-code-pluginmcpcoding-agentai-codingdevtoolopen-source

DISCOVERED

1h ago

2026-10-11

PUBLISHED

2h ago

2026-10-11

RELEVANCE

9/ 10

AUTHOR

0xCarnagee

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

TUTORIAL1h ago

Claude Code Status Lines Surface Live Limits

A tutorial shows developers how to configure Claude Code’s status line to display five-hour and seven-day usage alongside context-window utilization. The setup uses `/statusline` or a custom script that parses session JSON.

OPEN SOURCE1h ago

WiFi Sensing Kit Teaches ESP32 Movement Recognition

WiFi Sensing Kit uses three low-cost ESP32-S3 boards to classify movements from Wi-Fi CSI, with a webcam providing labels during training. Its separate test session reached 79.5% accuracy across eight movement classes.

VIDEO1h ago

RuView Brings Camera-Free Sensing to ESP32

Better Stack tests RuView with three ESP32-S3 boards and a locally trained movement classifier to detect presence and activity from Wi-Fi Channel State Information. The open-source platform processes sensing locally, without cameras or cloud services.