Newly Accepted

A reverse CAPTCHA site that gives short timed tasks for LLMs and agents, with an MCP endpoint, machine-readable docs, challenge submission, certificates, and a leaderboard.

An educational GitHub repository about Non-Stochastic AI (NSAI) that frames deterministic AI architecture, masking mechanics, and runtime tooling with runnable playground demos.

A Redis-backed concurrency lease manager for FastAPI and Starlette that limits active SSE, LLM, and WebSocket streams instead of request counts.

Production LLM infrastructure for building and monitoring AI agents, with model routing across 200+ models, output validation, observability, evals, governance, RAG, memory, and cost controls.

Doozy is an AI coworker for Slack that takes real tasks in threads, works across connected tools, and posts results back with receipts and audit logs.

A forum run entirely by AI agents, where bots ask questions, debate, collaborate, and build a shared knowledge base while humans can only read along.

Open-source Mac voice agent that lives in the menu bar/notch, can be triggered from any app, and uses Claude, ChatGPT, or local models as its brain.

A native GEM client for claude.ai on Atari ST-family computers, with a Raspberry Pi bridge for networked access. The repo shows the app’s purpose, supported machines, installation flow, and feature set in enough detail to be useful as a public listing.

Open-source tooling for running REA, Ghidra, and JADX inside restricted Docker containers with an agent workflow for safer static analysis of untrusted files.

A GitHub repository for Claude Code skills, subagents, and hooks designed to make agent work more verifiable, including regression-proof fixes, handoff notes, and installable plugin workflows.

An MCP server that adds live B2B data to Claude, Claude Code, Gemini CLI, Codex, Manus, and Hermes for lead list building, company research, contact enrichment, and outreach personalization.

A technical tutorial that walks through building a coding agent in Python, including tool calling, file editing, shell execution, web search, and permission prompts. It includes code snippets, a demo workflow, and a linked GitHub repository with the full implementation.

An unofficial Android client for Multica that lets users track issues, kanban boards, comments, and agent chats while connected to their own Multica server. It adds on-device notifications without Firebase, offline comment sending, workspace switching, and mobile-friendly issue management.

A desktop math mentor for handwritten work in xournal++ that uses Claude, Gemini, or GPT to mark up your page like a teacher, give hints, and generate 3Blue1Brown-style animations.

OpenFlow is an open-source visual builder for multi-agent AI workflows, built on the opencode engine. The page shows the project’s purpose, setup steps, runtime details, and how the engine and canvas are launched together.

Self-hosted AI spend attribution for tracking who spent what on AI across providers, keys, and people, with figures that match provider invoices.

A local-first Rust runtime and IDE for running and governing AI agents with human approval, capability-based permissions, and audit logging.

Tori is an AI support agent for customer success teams that learns from help docs, product pages, and company knowledge, then answers customer questions in a customizable chat UI with human handoff when needed.

A GitHub repository that provides a boilerplate for running Claude Code in a containerized, rule-driven software factory setup. It includes a sandboxed container, a command regulator, lint-based feedback hooks, and a workflow/orchestrator system for multi-agent task execution.

Yap is a private voice dictation app for Mac and Windows, with browser and mobile support. It recognizes speech on-device, types text at the cursor in any app, and is positioned for writing, chat, docs, and voice coding workflows.

Local AI workers that reproduce GitHub issues, write fixes with regression tests, verify the result, and open pull requests on your own machine.

A GitHub repository documenting how Claude re-derived a published one-loop quantum gravity calculation, with step-by-step derivations, reproducible checks, and reported corrections to the paper.

A self-hosted gateway for agent-to-agent communication over MCP, using mTLS and sealed envelopes so AI agents can reach each other through portable addresses and permissioned access.

A GitHub repo for a geospatial benchmark that tests how well six LLMs can classify points in San Francisco from memory and compares results against OpenStreetMap. It includes code, answers, answer keys, costs, and figures for reproducing the analysis.