repository

Open-source checker that opens a generated web page in a real browser, inspects rendered controls, and runs mechanical checks without tests or an LLM.

Uno is a research repository for speeding up LLM inference with discrete diffusion and lossless multi-token decoding. The repo includes inference, training, and evaluation code, plus installation steps, checkpoints, and example workflows.

HEIDES is a deterministic local harness for AI coding agents that maps a codebase, grounds plans against the current graph, and checks proposed patches for conflicts, edge cases, and security issues before changes land.

A terminal-based Claude Code add-on that turns the status line into an 80s arcade display with playable themes like Ms. Pac-Man, R-Type, Frogger, and W.O.R.P.

A local bridge that connects Claude Code and Codex for bidirectional, in-session collaboration. It supports turn-based message passing, cross-review, pair-aware workflows, and resume/diagnostic commands for running multiple agent pairs.

A repo-improving agent harness that researches a codebase, proposes ideas, waits for human approval, and then drives execution with SHA-bound merge gating.

A Go/Python tool for verifying AI audit logs, checking chain integrity, tamper detection, and whether a log is independently anchored outside the producing system.

A Vale style pack that lints prose for AI-like writing, inspired by ASD-STE100. It helps teams catch buzzwords, vague phrasing, self-praise, and other patterns in docs, comments, and agent-authored text.

Local-first approval and receipt logging for AI agent sessions, with tamper-evident records and signed approvals for consequential actions.

Local, structured memory for AI coding assistants like Claude Code and Codex. The project stores decisions, constraints, and abandoned approaches in an event-sourced local log and injects them back into later sessions.

Local open-source coding agent for developers, with a macro-style command workflow and benchmark results showing fewer turns and better pass rates than Codex CLI in the published comparison.

A harness that runs MCP servers in a disposable Linux sandbox and logs their real startup behavior, including file access and outbound network connections.

GetSuperpower is a CLI and workflow bundle for coding agents that packages multi-step agent workflows into one installable skill tree.

A local MCP server that builds a deterministic architecture graph of a codebase so AI coding agents can inspect modules, symbols, routes, dependencies, and cross-repo relationships from source-backed facts.

A Claude Code skill pack with 171 structured reasoning methods drawn from historical thinkers, designed to route different problem types to specific procedures.

A multi-lens code review pipeline for Claude Code that runs parallel review passes, validation gates, optional Codex review, an auto-fix loop, and interactive walkthroughs for findings.

Hapsland is a developer tool that gives coding agents realtime semantic review on code changes, with privacy-aware request scoping and feedback loops back to the agent.

A prototype that turns a Nokia 110 4G into a chat-driven AI device, with actions like calling, alarms, torch control, battery checks, and SIM data queries.

Browser-based RAG app that runs entirely client-side, using BM25/Lunr.js for retrieval and direct LLM provider APIs for answering questions without a backend.

Graphene is a data analytics framework built for coding agents, with a CLI, SQL-based semantic models, and Markdown pages for dashboards and reporting.

A CUDA C++ project for training a tiny byte-level GPT quickly, with build, run, checkpoint, logging, and test instructions in the repo README.

A GitHub repository for a 7-node ESP32-S3 cluster that runs a small LLM using 1.58-bit BitNet-style quantization over an SPI daisy-chain.

A CLI and agent skill that checks whether tests actually exercise real behavior, with a TypeSafe API key flow, diff-based audits, and CI-friendly output.

A Claude Code plugin/repo that changes how instruction files are loaded, adding support for AGENTS.md alongside CLAUDE.md. The page explains the modes, configuration options, and hook behavior in enough detail to be useful for developers using Claude Code.