Recently Evaluated

Self-hosted runtime for putting hard budgets, approval gates, observability, and resumable checkpoints around AI agents.

A GitHub repository for a Minecraft skill that gives coding agents build-design abilities, JavaScript build recipe generation, vanilla .nbt export, and interactive previews before placing structures in-game.

dbmachine turns a Postgres database into a self-describing backend that AI agents can operate through a CLI and YAML spec. It compiles schema, constraints, RLS, audit logging, and custom operations into a local Postgres setup.

Chrome extension that brings AI coding help directly into the browser for debugging, code explanations, security reviews, and solution generation.

Go-based multi-source search CLI for LLM agents with session-based workflows, paginated result reading, and support for web, HN, Reddit, GitHub, YouTube, and multiple search APIs.

VaultS3 is a self-hosted, S3-compatible object storage server with a built-in dashboard, low RAM usage, and a single-binary deployment. The page highlights features like versioning, WORM, S3 Select, FUSE mount, IAM/OIDC login, search, and event notifications.

SecretSpec 0.12 adds audit logs and reason-required secret access for coding agents, with CLI, SDK, and config support.

Local-first agent OS that lets you run and steer Claude Code, Codex, and OpenCode from a browser or chat app while keeping code and keys on your own machine.

A font project and keyboard firmware concept that turns typed text into a noise-like character stream while rendering readable output only through the font itself.

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 custom Claude Code status line script that shows workspace, model, session, context usage, token totals, code churn, and 5-hour/7-day rate-limit runway in two colored rows.

An interactive, sourced deep-dive showing how Stack Overflow’s Q&A format helped train coding-focused AI systems and what happened as AI reduced public question volume.

GitHub Marketplace app from NeuraLegion that scans apps and APIs for vulnerabilities, proposes fixes, and validates remediations inside GitHub workflows.

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

A macOS desktop shell where apps are generated by Claude into interactive HTML mini-apps, with local system features like windows, dock, file storage, notifications, and persona-based themes.

An open-source shim that lets Cursor canvas .tsx files render on the web by aliasing cursor/canvas to a Mantine-backed implementation. Includes install steps, runtime setup, supported API coverage, limitations, and a live example deployment.

A cross-platform desktop app that turns raw screenshots into polished visual assets, with local privacy redaction, OCR, and an AI bug agent for reports.

Aquifer is a self-hosted MCP gateway runtime for buffering spiky agent and API traffic. It queues requests durably in SQLite, dispatches them at a configurable rate, and supports both MCP and HTTP adapters with webhook delivery.

AI sales agent for cold outreach that finds prospects, personalizes messages, and follows up across email, text, LinkedIn, and X until someone replies.

An arXiv paper on reducing LLM inference cost for web automation by compiling browser tasks into a deterministic JSON workflow and executing them without repeated model calls.

A CLI tool that shares Claude Code sessions across a team by storing transcripts and memory on an orphan Git branch, so teammates can pull, resume, push, and delete sessions without extra infrastructure.

Lathe generates multi-part, hands-on technical tutorials from prompts and provides a local UI plus CLI workflow for working through them with LLM-assisted skills.

An interactive guide and demo for prompt engineering that shows how analogies, examples, and format-matching can improve LLM outputs.

A minimal native macOS sandbox for running Claude Code or Codex with OS-enforced limits on what the agent can touch.