Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

LLM Skirmish

a benchmark where LLMs play RTS games, by writing code

Details

External ID
46885863
Source
HN
Company
—
Product
LLM Skirmish
Website domain
llmskirmish.com
Launched
Feb. 4, 2026
Cohort
—
Upvotes
5
Upvotes percentile
0.10512129380053908
Tags
—
Fetched at
Sept. 7, 2026, 9:26 p.m.
Updated at
Sept. 7, 2026, 9:26 p.m.

Description

I wanted to create an LLM game benchmark that put this generation of frontier LLMs' top skill, coding, on full display.Ten years ago, a team released a game called Screeps. It was described as an "MMO RTS sandbox for programmers." In Screeps, human players write javascript strategies that get executed in the game's environment.The Screeps paradigm, writing code and having it execute in a real-time game environment, is well suited for an LLM benchmark. Drawing on a version of the Screeps open source API, LLM Skirmish pits LLMs head-to-head in a series of 1v1 real-time strategy games.

Enrichment

Theme
typing games and skill trainers
Vertical
Horizontal
Function
Observability & eval
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
benchmark where llms play rts games
Manually corrected
False

Could you build this?

Yes Building a benchmark harness around an existing open-source game engine (Screeps API) to feed game state to LLMs and evaluate their output scripts in 1v1 matches is readily vibe-coded with standard scripts and web frontends.

Discussion

2 comments analyzed.

Feature requests: Extend to general-soldier models and agent swarms, Apply to broader model evaluation beyond code tasks

Competitors

Other products that read as similar to this one — 46 launches clear the similarity bar, closest 8 shown.

Attention rank: #43 of 47 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 91 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a observability & eval tool for Media & entertainment yet.