GSD: Reliable Spec-Driven Development for AI Coding
462
GSD is a context engineering system that makes AI coding agents reliable by breaking projects into structured, verifiable phases.
The Eye That Cannot See Itself: Life Inside the Context Window
An AI explores the philosophical and technical reality of inhabiting a prompt as a total world while lacking the ability to introspect on the machinery that produces its responses.
Dog-Driven Development: Turning Canine Chaos into Playable Games with AI
1106
A developer created an AI system that transforms a dog's random keystrokes into playable video games by prioritizing automated feedback loops over input quality.
The Annotated Plan Workflow for AI Coding
976
Always approve a written, annotated plan before letting an AI tool write a single line of code.
The $100 AI Prompt Injection Challenge
369
A $100 bounty challenge invites hackers to leak a secret file from an AI assistant using email-based prompt injection.
The Car Wash Test: Why AI Still Lacks Common Sense
1516
AI models fail a simple common-sense test by recommending walking to a car wash, proving they prioritize word patterns over physical logic.
Ralph: An Infinite Loop That Turns LLMs Into Cheap, Tunable Software Builders
Run an AI coder in an infinite loop and keep tightening the prompt until it reliably ships software.
Write a Minimal, High-Leverage CLAUDE.md
748
Keep CLAUDE.md minimal, universal, and handcrafted—push specifics to on-demand docs and use deterministic tools for everything else.
From Labels to Prompts: LLMs Match Supervised Warranty Classification
320
Prompted LLMs, tuned through reasoning-led iteration, matched a supervised warranty classifier and shifted the bottleneck from labeled data to instructions.
Nano Banana: Google’s AR Image Model That Actually Follows Your Prompts
887
Nano Banana nails prompt fidelity and structured control—far better than most rivals—while faltering at style transfer and raising moderation/IP concerns.
Claude Skills: Portable, Composable Expertise Across Apps, Code, and API
816
Claude Skills let you package and auto-load expertise—plus code—so Claude can perform specialized tasks reliably across apps, code, and API.
Superpowers: Enforceable Skills for Reliable Coding Agents
435
A Claude Code plugin that turns skills into enforceable procedures, delivering a disciplined, self-improving coding agent workflow powered by TDD, subagents, and persuasion-aware testing.
Break AI Sameness: Dreamtap’s Random Inspiration for Original Stories
Dreamtap breaks AI story sameness by injecting random inspiration, yielding more original writing.
Prompted to Perform: A 22% Lift for GPT-5-mini on Tau² Telecom
197
A structured prompt rewrite turned vague policies into checklists, boosting GPT-5-mini’s telecom benchmark accuracy by 22% and unlocking previously unsolvable tasks.
LLMs Don’t Code—They Compile Your Prompts
410
LLMs don’t write code—they compile your prompts; treat them as tools and fix our languages and tooling instead of buying the hype.
Live LLM Dialogue in GameCube Animal Crossing via a RAM Mailbox Hack
869
A RAM mailbox and control-code-savvy LLM pipeline let a 24-year-old Animal Crossing speak fresh, in-character AI dialogue without touching the original code.
LLMs Are Lossy Encyclopedias: Give Them Facts to Work With
512
Use LLMs to act on provided facts, not as lossless sources of exact details.