Gemini 4 Argon plays StarCraft
Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.
Read the full description
I've refreshed StarSkirmish Bench, a benchmark where LLMs play StarCraft BW by writing code, with results from the flagship models of OpenAI, Anthropic, Google, and Muse
Visit website ↗ Source listing ↗
HN · launched Oct 16, 2026 · 254 upvotes (top 2%)
- Media & entertainment
- Agent / copilot
- Prosumer
- AI-native
- Hobby / open-source project
Similar launches
See all 41 similar launches →Could you build this?
Partial The evaluation harness that runs automated StarCraft: Brood War matches and logs scores can be assembled with scripts, but orchestrating the BWAPI environment, sandbox execution, and dynamic C++ bot compilation at scale is non-trivial.
What it would actually take: Requires setting up a Linux/Wine or Windows virtualization harness running BWAPI (Brood War API) and StarCraft headless instances in isolated Docker containers. The hard part is building an automated, sandboxed agent evaluation loop that securely compiles untrusted C++ LLM code on the fly, monitors frame-by-frame match performance via BWAPI, and aggregates statistical tournament ratings (Elo/Glicko) reliably across parallel game runs.
Discussion
20 comments analyzed.
Competitors mentioned: broodbench.com
Concerns raised: Bots might rely on memorized open-source bot code, 1-hour code generation limit is incoherent, Unclear what end-to-end benchmark measures in what proportions
Feature requests: Benchmark on Red Alert 2 / Red Alert, Benchmark on Cities Skylines, Include charts for all 45 LLMs like GLM
Competitors
Other products that read as similar to this one : 41 launches clear the similarity bar, closest 8 shown.
Attention rank: #2 of 42 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).
Launched 309 days after the earliest competitor.
- argon-games · github · 2026-10-13 · 31 upvotes · similarity 0.52
- Gemini 4 Argon · ph · 2026-10-04 · 289 upvotes · similarity 0.46
- 15 Years of StarCraft II Balance Changes Visualized Interactively · hn · 2026-01-10 · 9 upvotes · similarity 0.43
- Watch LLMs play 21,000 hands of Poker · hn · 2026-01-08 · 36 upvotes · similarity 0.43
- awesome-gemini4-argon-videos · github · 2026-10-13 · 17 upvotes · similarity 0.43
- agy-argon-bridge · github · 2026-10-14 · 15 upvotes · similarity 0.42
- LLM Skirmish · hn · 2026-02-04 · 5 upvotes · similarity 0.42
- I made a hockey web game with Opus 5.5, and Gemini Argon optimized it · hn · 2026-10-13 · 5 upvotes · similarity 0.40
Other launches for this product
- No other launches for this product.
More details
- External ID
- 50118688
- Company
- n/a
- Product
- Gemini 4 Argon plays StarCraft
- Website domain
- starskirmish.com
- Cohort
- n/a
- Upvotes percentile
- 0.98
- All tags
- n/a
- AI summary
- ai agent playing starcraft
- Manually corrected
- No