Top Picks

A Claude Code skill that rewrites Claude’s responses through Gemini to strip out the over-polished, BuzzFeed-like tone and output plainer English.

A DeepSeek Harness plugin that uses an LLM to verify and rank candidate outputs with select, compare, track, and rollout tools.

Vomit is a Go-based CLI that converts Claude output into readable English by sending it through a separate local LLM. It supports install/setup commands, a scrub mode for Claude hooks, and sidecar commands for listing and tailing sessions.

A pricing analysis article comparing batch API discounts across Google, OpenAI, Anthropic, xAI, and OpenRouter, with a focus on where batch rates differ from standard pricing.

An MCP server that converts images into ASCII art so text-only LLMs can attempt image interpretation. The README includes a short explanation, example setup, and a clear description of its limitations.

HyperSAE is a Python package for mechanistic interpretability that trains hyperbolic sparse autoencoders on LLM activations. The page includes installation steps, a quickstart example, benchmark tables, and a short software architecture overview.

Aakit is a developer tool for measuring and tracking assumptions made by coding agents, linking each assumption to supporting code and flagging which ones fail. The repo includes a clear explanation of the idea, prior art, and early experiment results.

A public showcase for Mangudai, an external attack surface management and vulnerability management platform for small security teams. The page describes architecture, scan modules, pricing tiers, and the intended SaaS workflow, making it suitable for a directory listing.

A CLI/library for line-level provenance in text edited by humans and agents, using version history to estimate which ranges are human-written versus machine-generated.

OneRingAI is an open-source TypeScript library for building AI agents with integrations and graph memory. The repository shows a substantial codebase with docs, setup files, testing, MCP-related material, and references to supported models and multimodal capabilities.

A technical case study on reducing token usage in a coding agent, including a model-free harness, wire-level measurement, and several concrete optimizations that cut a long run from 1.04M to 517k tokens.

An open-source project for running a small LLM and limited agent workflows on ESP32-P4 hardware, with model details, architecture notes, examples, and setup instructions.

Market-derived RL environments for training agents on quant trading and long-horizon planning under adversarial noise.

A standalone CLI coding agent from Warp that runs in any terminal and supports multi-model routing, persistent sessions, full-screen terminal apps, and cloud-agent handoff.

A searchable library of 71,262 hand-drawn SVG icons with AI/MCP access for tools like Claude Code, Cursor, VS Code, and Codex.

An open-source multiplayer agent harness for work, designed for Slack and web-based collaboration with scoped memory, permissions, sandboxes, and admin controls.

A research page on how censorship and behavior transfer during model distillation, with published models, data, evaluation code, and a benchmark called LineageEval.

A Claude Skill that rewrites technical text into ASD-STE100 Simplified Technical English. The repo includes before/after examples, install commands, and references for using it with tools like Claude Code and GitHub Copilot.

A web app for branching AI conversations into a DAG-style chat tree, with model switching, response style controls, chat import/export, and example conversations.

A GitHub repo package about using ASD-STE100 Simplified Technical English to reduce AI-generated writing slop, with a distilled writing skill, a heuristic linter, and experiment results across models.

A repository of reusable AI agent skills centered on regression evals and benchmark history, with install instructions and local eval workflows.

A local speech-to-text MCP server for AI agents, with CLI transcription and support for Parakeet or Granite backends.

Foomchat is an AI chat that generates installable UI features from a user prompt. The page shows a simple workflow: describe a feature, install it into an app, and share it through a curated marketplace.

A Codex Skill that installs verified public-safe skin packs from GitHub releases, validates theme files, stages them locally, and shows rollback guidance before applying anything.