What is mattwood.fyi?

A live, structured feed of riffs and links from Matt Wood, Chief AI & Technology Officer at AWS — readable by people and queryable by agents as a typed knowledge graph.

Is there an API and is it free?

Yes. A public, read-only query API is available with no authentication and no API keys. See https://mattwood.fyi/openapi.json and https://mattwood.fyi/docs.

How should an AI agent use this site?

Fetch https://mattwood.fyi/feed.json for the full structured feed, or query the API for keyword/semantic search and the connection graph. See https://mattwood.fyi/agents/ for when-to-use guidance.

Your Game Is A Harness — mattwood.fyi

Connections

Related to

Thinking Before Thinking: Scaling Agentic Inference Through Meta-Reasoning Both address how agentic AI systems should reason and iterate within constrained environments - meta-reasoning for scaling inference complements game-legibility design for accelerating agent development cycles
Unreal Agent: Cost-Efficient AI Agent Harness Both use the 'harness' metaphor for AI agent environments - 'Unreal Agent: Cost-Efficient AI Agent Harness' directly parallels the concept of designing systems (games/frameworks) as structured environments for AI agent operation
Coding Agent Continual Learning: Analysis of ARC-AGI-3 Problem Solving Both examine how AI coding agents iteratively solve structured problem environments - the game-as-harness framing parallels ARC-AGI-3's use of puzzles as structured iteration surfaces for coding agents

Supports

Can LLMs Play NetHack? A 2026 Agent Benchmark Both explore LLMs operating within game environments as agents - 'Your Game Is A Harness' provides design principles that would directly inform the kind of benchmarking approach used in 'Can LLMs Play NetHack?'
Automating Eval Design and Hillclimbing with Claude 'Automating Eval Design and Hillclimbing with Claude' shares the core methodology - using structured environments (evals/games) with Claude to enable iterative AI improvement loops
An LLM Beat NetHack 'An LLM Beat NetHack' is a concrete example of the thesis in 'Your Game Is A Harness' - that games with legible state/action spaces enable successful AI agent iteration

Develops into

StarSkirmish Bench v0.2: LLM StarCraft Bot Benchmark StarSkirmish operationalizes the 'game as harness' concept - using StarCraft as a rigorous evaluation harness for LLM coding ability with measurable win/loss outcomes
Object-Oriented Harness: Evolving Agent Harnesses through OO Design Both treat harnesses as first-class design artifacts; OO Harness extends the philosophical framing of 'your game is a harness' into a systematic evolutionary methodology

Supported by

StarSkirmish Bench v0.2: LLM StarCraft Bot Benchmark StarSkirmish treats the game (StarCraft) as a harness for evaluating LLM coding agents, directly embodying the 'your game is a harness' concept where game mechanics serve as evaluation infrastructure

Challenged by

Object-Oriented Harness: Evolving Agent Harnesses through OO Design OO Harness treats agent harnesses as optimizable software artifacts evolved through design principles, while 'Your Game Is A Harness' frames harnesses as structural constraints - the former actively subverts the static harness conception

Type:	link
First published:	Tuesday, 13 October 2026 at 17:54 UTC
Location:	https://kylekukshtel.com/ai-harness-engineering-game-development-claude-code
URL:	https://mattwood.fyi/i/dc7befd6

I'm Matt Wood, and this is For Your Information. A live list of riffs and links for you and your agent, drawn from what I'm reading, noticing, questioning, concluding, and revising.

Connections