developer tools

A research note from OpenShell on using formal methods and Z3 to reason about and verify permission changes in long-running AI agent systems.

GitHub’s research preview on multi-model orchestration for coding tasks. The page describes adaptive workflow selection and includes benchmark comparisons against the Opus 5 baseline with estimated cost reduction claims.

Oconee Runtime is an enterprise AI governance product focused on policy enforcement for browser AI and coding agents. The page explains how it evaluates proposed agent actions against policy and supports allow, warn, and block outcomes.

An educational project that explains and demonstrates a homemade language processing unit inspired by Groq’s LPU, with a focus on running a small Transformer model and visualizing the idea behind AI hardware.

A native Emacs mode for chatting with AI agents via ACP, with prompt queueing, chat mode, image handling, and buffer navigation improvements.

A technical blog post from the Cua repository describing a process-scoped Metal capability shim for macOS VMs that improves llama.cpp performance on Apple Silicon, with benchmark results and reproduction details.

Open-source Python project for execution-first multi-agent workflows built around YAML Charterfiles, runtime schema enforcement, and a boss-agent hierarchy. The README includes install commands, a short usage tour, and a small code example.

A playable text-adventure that teaches Claude Code through real prompts, skills, and agent workflows.

A Vite plugin that obfuscates client-side code in a way meant to make LLMs refuse de-obfuscation attempts. The README shows the project’s purpose, a usage note, and an example of the transformed output.

A RequestRocket article mapping the AI agent access-control stack across authentication, authorization, governance, identity, gateways, and runtime egress control. It explains how the categories differ, when each fits, and where RequestRocket sits in the landscape.

An in-depth Medium post from IronBee about how a verification loop affects AI coding agents, using Web-Bench and comparing DeepSeek with Claude Opus on a real web app task.

A macOS terminal workflow that uses local speech-to-text plus the Pi coding agent to turn spoken requests into shell commands or terminal answers.

Anthropic research report on how people use Claude Code in practice, based on a large privacy-preserving analysis of session data. It covers task types, division of labor between user and model, and how domain expertise affects outcomes.

A blog post describing an evaluation harness for comparing agentic coding tools and prompts across realistic SWE tasks, with token-cost results and model-specific behavior notes.

A shareable report for AI coding usage that summarizes how you work with tools like Claude Code, Codex, and Cursor, with sign-in via GitHub and an option to get printed copies.

LEVI is a harness-first evolutionary framework for code and prompt optimization. It focuses on reducing LLM cost with diversity-preserving search, role-aware model routing, and a proxy benchmark, and presents comparative results against several existing systems.

A blog post describing a small reinforcement-learning agent trained with PPO to play and beat a Pokelike/Pokerogue-style game, including the input representation, model architecture, and training approach.

A Codacy blog post explaining GitHub Copilot’s new AI Credits billing for code review and positioning Codacy AI Reviewer as an alternative with fixed per-seat pricing, static analysis, and pull-request review features.

A tutorial showing how to build a release notes generator with the Cline SDK, including project structure, setup, custom tool design, and code walkthroughs.

A detailed article about the internal harness Nimbalyst built around Claude Code and Codex, covering context, provenance, capability, workflow, restraint, verification, visual interface, and coordination.

A minimal LLM agent packaged as a single HTML file. The README explains how to run it locally, set the LLM host, API key, and model, and use chat("hi") from the browser console.

A Rust CLI that wraps Claude Code for programmatic use, forwarding flags, injecting one-shot prompts, tailing session transcripts, and exiting automatically after the turn completes.

A Python pipeline that ranks 1,000 Show HN posts with an LLM judge and TrueSkill, then compares estimated merit against Hacker News points.

A merged Visual Studio Code pull request that changes the Git extension so AI co-author trailers are enabled by default. The snapshot includes the configuration change, the affected files, and Copilot review feedback.