open-source

A source-available JavaScript browser agent for asserting end-to-end behavior in real user sessions. The repo explains the annotation-based approach, how it runs in staging and production, and how to install and initialize it via CDN or npm.

A sourced model and short report exploring whether Europe could train a sovereign frontier AI model using public compute it already owns, with reproducible code, datasets, and a PDF report.

OpenLoomi is an open-source desktop AI workspace that connects to apps like email, calendar, chat, docs, and project trackers to build a persistent context graph with proactive actions and local-first storage.

An Emacs package that acts like an AI pair programmer and teaching assistant while you code, with save-time reviews, follow-up questions, and architecture memory for local projects.

Cast is an open-source, self-hosted harness for multi-user Claude agents. It focuses on routing, identity, and access control outside the prompt, with support for agent building, review flows, and extensions like email, calendar, web fetch, and WhatsApp.

A technical guide that shows how to build a small CLI-based LLM agent for microservices using tool discovery, model calls, conversation history, and a REPL loop.

Smriti is an open-source system for shared reasoning state across coding agents like Claude Code and Codex. It treats project reasoning like version-controlled state, with local setup, a CLI, backend, and frontend for coordinating multiple agents on the same repo.

Open-source AI SEO optimizer CLI that audits pages, generates meta tags, analyzes keyword gaps, and drafts blog outlines. It can run free for audits and uses a KianBot API key for AI-assisted features.

An open-source research framework for structured LLM evaluation, claim verification, and source-grounded reflective reasoning. The repo describes modular components for retrieval, semantic scoring, skeptical claim checking, and benchmark-style epistemic assessment.

A blog post documenting an open-source, under-$40 take on OpenAI’s Codex Micro, built as an AI-assisted PCB routing and keyboard/macropad project.

MindShield is a browser extension that adds a reflection pause before sending simple prompts to AI tools, using local on-device classification to reduce false positives and keep prompts private.

Local, private CLI that reads Claude Code session transcripts and estimates API-equivalent usage, token counts, and cost, with optional ledger export.

Canonical Agent Notation (CANON) is an open-source controlled English for model prompting and agent-to-agent communication. The repository describes a small closed vocabulary, mandatory obligation markers, bounded loops, and a clear ask-rather-than-guess rule for making requests and agent messages less ambiguous.

A browser extension for Chrome and Edge that connects a local folder to claude.ai so Claude can read, search, and edit files without a desktop app or API key.

A lightweight TypeScript resilience layer for LLM calls with retries, timeouts, caching, circuit breaking, typed structured output, and usage tracking across many provider clients.

Hosted server and open-source repo for playing multiplayer games with Claude Code in the CLI, starting with chess and comedy battle modes.

An open-source, Kubernetes-shaped control loop for agent-driven work. It watches external systems, makes one LLM-driven move per beat, and supports human-in-the-loop approvals through a file-based contract.

taste-ai is a command-line tool that compresses project context for AI agents and learns coding patterns from session logs, git history, and prompts. The README shows install steps, a quick start, command examples, and a workflow for generating compact session context and updating pattern files.

Nodea is a live branching AI chat canvas for working through complex projects. The demo shows forkable replies that turn conversations into trees, with example prompts for naming a coffee shop and shaping branding ideas.

A CLI tool that keeps Claude Code, Codex, and GLM 5-hour rate-limit windows continuous by auto-pinging providers when a window resets.

kLex is the reference runtime for FROG, a typed language built for native tooling, browser apps, pipelines, and concurrent utilities. The repo shows install steps, language examples, a browser playground, GUI support, and a security-scanner demo.

StackNudge is an open-source notification tool for AI coding agents. It adds banners and sounds when supported agents finish a task or pause for approval, with macOS support for a native panel and click-to-focus behavior.

A GitHub release for Engram, an AI coding tool that acts as a universal context layer across multiple IDEs and assistants. The v3.4.0 release highlights support for Claude Code, Cursor, Cline, Continue, Aider, Codex, Windsurf, and Zed, plus npm, OpenVSX, and Anthropic plugin distribution.

CtxVault is an open-source receipt layer for governed AI context. It shows what evidence was selected, caveated, blocked, omitted, and how a context packet can be audited or rolled back.