Sandboxed harness for testing coding-agent learning

Arc Code tests whether a general coding agent can learn unfamiliar ARC games by creating its own parsers, simulators, and search programs during play.

From Github Awesome — GitHub Trending Weekly #45: comet, blobatar, cumora, herdr, barehands, ip-as-logo-skill, Aura, md2hd at 06:27

Problem: Agent evaluations need to test tool creation and adaptation rather than only predefined task execution.

For: Researchers evaluating general coding-agent capabilities.

Products from this video

ai-data-extractor AI Design Skills APEX Inference Chip arc-code Aura barehands blobatar Core Framework Cumora DesktopFly dgit Endoplexity herdr HQBase icm-architect IP as Logo Skill jit macOS Harness md2hd morphnext NorthCinder Procedural Sounds sloptrim TrueForge unlazy Vibe ASO VibePulse Viscose Zeron

Examples

🔒 Full analysis locked

Unlock more videos and the full analysis

A credit unlocks one video's full analysis for good — the build steps, the tools and how each was used, the methods behind every use case. Pro opens the whole library instead, and raises how many videos you can analyse a day.

Unlock full analysis — free

Related ideas