Back to Project
S

StarSkirmish

Project Oct 16, 2026

A StarCraft Brood War coding benchmark giving language models one hour to build C++ Protoss bots, with v0.2 evaluating five runs against a fixed opponent roster.

Visit primary link 2 sources attached
StarSkirmish's published game montage poster showing Skynet versus a GPT-6 Sol bot on Heartbreak Ridge in StarCraft Brood War.
Poster frame from the organizer's published Bench game montage, visibly labeled Skynet and GPT-6 Sol. It illustrates bot gameplay, not a reproduced AITrending match or proof of the v0.2 aggregate scores. Full frame preserved. Source: StarSkirmish / Kai McPheeters; StarCraft by Blizzard Entertainment
ai-agentsagent-ecosystem

What happened lately

StarSkirmish timeline

StarSkirmish tests agentic coding through StarCraft: Brood War bot development. Models write C++ against BWAPI 4.4.0 and test their bots using OpenBW. They do not directly control the game through visual actions.

Version 0.2 runs models in their labs’ first-party coding CLIs, with five one-hour attempts per model and a common roster of human-written and demonstration bots. The published scores are the organizer’s evaluation results.