I build AI platforms, agent tools, and web apps focused on trust, storytelling, and automation.

github

Joshua Herron  • Peerlist

I built Infinite Realms, an AI-powered text-based roleplaying game where players can create a character, enter different worlds, and continue evolving across sessions through persistent memory and campaign state. The platform is designed around dynamic storytelling, evolving characters, long-term progression, and central storylines that respond to player choices. Instead of resetting after each session, Infinite Realms keeps track of character progress, objectives, inventory, relationships, reputation, and world state so the experience feels like a living RPG campaign. I built the platform using Next.js, the Vercel Agent SDK, Supermemory, Drizzle ORM, PostgreSQL with Neon, Python, Modal, ComfyUI, and OpenRouter. I handled the full product development process myself, including the frontend, backend, database structure, AI orchestration, persistent memory systems, world simulation, character systems, and generative storytelling pipeline. Infinite Realms is the first phase of a larger vision: building toward a persistent AI-driven RPG world that can eventually expand into more immersive experiences.

Infinite realms

I built NovelHut, a digital publishing platform for writers and readers, completely by myself. NovelHut is designed to give authors a place to publish their books, organize chapters and volumes, grow an audience, and interact with readers through reviews, ratings, comments, and personal libraries. Readers can discover stories, follow books, leave feedback, and engage with authors in a community built around serialized fiction. As the solo developer, I handled the full platform from idea to launch, including the frontend, backend, database structure, authentication, publishing tools, reader experience, author dashboard, search, content organization, and core community features. This project taught me a lot about building a real product end to end, from designing user flows and database models to shipping production features and improving the platform based on how people actually use it. Website: https://www.novelhut.net

Novelhut

I built OpenTrust, an open-source trust and identity layer for AI agent tools. OpenTrust is designed to give AI agents a standardized way to understand whether a tool is safe to use before calling it. The platform uses “Agent Tool Passports” to declare a tool’s identity, permissions, trust level, version history, source formats, and review status across ecosystems like MCP servers, OpenAI functions, LangChain tools, OpenAPI endpoints, npm packages, and PyPI packages. I built the full system myself, including the passport schema, registry API, web frontend, CLI, SDKs, badge generator, manifest validator, payment contract interfaces, and the Hands Body and Feet MCP capability server. The system includes trust-level enforcement, permission scoping, signed passports, revocation support, spend caps, kill switch controls, GitHub identity verification, wallet verification, and machine-readable payment metadata. OpenTrust was built around a simple belief: as AI agents gain the ability to access files, wallets, APIs, terminals, infrastructure, and real-world services, they need a trust layer that is open, verifiable, and framework-agnostic.