AI Tools — leaderboards.

W

WhisperBar Trying to Fix Both Reading and Writing in the AI Age

WhisperBar streamlines AI interactions by providing rapid, reliable voice transcription alongside enhanced text comprehension tools

F

Forgot the Command? Stay in the Terminal

HumanSH is an in-terminal AI assistant that translates plain English prompts directly into executable shell commands

B

BAIhAIs

BAIhAIs is a multi-agent simulation where autonomous AI residents experience an ongoing art school ecosystem

D

Derive

Derive is an open platform dedicated to hosting, sharing, and orchestrating AI workflows and artifacts

O

Open Session, the open

Open Session is a self-hostable agent orchestration framework designed to run multi-agent workflows across diverse AI models

B

BirthdayVideoAI

BirthdayVideoAI is an automated video creation platform designed to craft custom birthday wishes and celebrations

K

Knowl

Knowl provides advanced memory infrastructure designed specifically for autonomous AI agents

O

One source of truth for agent configs across repos

The Agency provides a unified hub to manage, version, and sync AI agent settings and prompts across distributed codebases

A

All my mail accounts in one read

Consolidate multiple email accounts into a single, secure read-only Model Context Protocol server

T

ThunderPhone v2

ThunderPhone v2 provides high-performance voice agents engineered for automating phone calls and powering conversational voice features in apps

b

bAIseball.org

bAIseball

S

Spore

Spore enables users to execute machine learning models locally on their own infrastructure while providing seamless remote access from any device

B

Backprompter

Backprompter streamlines AI application development by eliminating backend, database, and secrets configuration

V

VidScio

VidScio is an AI-powered discovery tool designed to find movie and television series titles from web links, images, or text descriptions

I

I let a subagent workflow refactor my codebase for three days

Oh My Subagents coordinates specialized AI subagent workflows to execute continuous, deep refactoring across entire software codebases

D

Diet Cola themed Chrome Extension to track you Claude usage

A browser extension designed to help users keep tabs on their Claude AI interaction limits and message quotas

A

A distributed agent runtime with SWIM mesh and zero

JarvisCore is an open-source runtime engine engineered for coordinating decentralized AI agents at scale

C

Crilio

Crilio provides a pytest-style testing framework designed specifically for prompt engineering

R

Reproduce a cross

A development and testing harness designed to simulate multi-agent handoffs across organizational boundaries

K

Keenable

Keenable provides a dedicated web search API powered by an index of over 100 billion pages

S

Searching for optimal strategies in a puzzle game with chance (Drop7)

An algorithmic research project and solver designed to determine optimal play in the chance-based puzzle game Drop7

C

Codex / Claude Code harness for Java high performance improvements

An MCP server and testing harness designed to bridge LLMs with Java execution environments

A

A hash

A hash implements the SDI protocol, a safety‑first framework that runs large language models one step at a time through a compile‑gate grammar

L

Local dictation and meeting notes for Mac to keep your data private

A Mac‑only app that transcribes speech locally, turning meetings into searchable notes without sending data to the cloud

I

I made a Raspberry with Qwen my local car AI

A 35B Qwen LLM runs locally on a Raspberry Pi, providing robust car diagnostics and control

A

A portable evidence record for what an AI agent ran

This tool creates a tamper‑proof log of every operation an autonomous AI agent performs, ensuring full traceability

S

Sillage a 4 MB memory that lets a frozen language>model remember

Sillage adds a lightweight 4 MB external memory to frozen language models, enabling them to retain context across interactions

D

Declaude

Declaude refines Claude's responses, stripping away the model's quirky phrasing for clear, human‑readable text

V

VinvAI runs and observes your services to find bugs and perf issues

VinvAI continuously monitors your applications in production, using AI to spot bugs, latency spikes, and resource inefficiencies

S

Stash, turn your Instagram saves into notes Claude finds on its own

Stash automatically transforms your Instagram saved posts into organized notes, using Claude to extract key insights

A

Active Source of Truth for Your Coding Agents

Meetless

D

Declarative, reproducible configuration materializer for AI agents

A declarative tool that lets you specify AI agent configurations in a reproducible format

E

Ever Wanted to Call Codex from Claude Code? My Harness Orchestrator

An open-source tool that lets you invoke Codex models directly from Claude code

T

Traccia

Traccia provides end‑to‑end observability for autonomous AI agents, letting you monitor actions, enforce runtime policies, and keep audit trails

H

Heimdall

Heimdall provides a trust‑verified knowledge layer that AI coding agents can query for reliable information

M

MiniMax H3 on a 16GB Mac, 5 days after open weights

MiniMax H3 demonstrates that advanced AI models can operate on a 16GB Mac within days of release

F

Find and Organize Photos with Private, Local AI

A privacy‑first tool that uses local AI to automatically tag and group your images

I

I turned Apple

A tool that converts Apple‑style scroll‑video webpages into AI‑driven skills you can reuse across platforms

A

A complete companion for agentic development

The Burmese platform offers a comprehensive suite for designing, testing, and deploying autonomous AI agents

A

Agent Office (Slack for AI Agents)

Agent Office provides a Slack‑like interface where AI agents can chat, share data, and coordinate tasks in real time

A

AgentSight

AgentSight provides real-time observability for autonomous AI agents using eBPF tracing

C

Caspian

Caspian provides a SDK that streamlines messaging between autonomous AI agents

S

Steganeur

Steganeur encodes hidden messages by steering an LLM's token choices, producing natural‑sounding cover text

P

Public Muscriptor Instance (latest, most powerful Audio

Pianoify leverages GPU-powered AI models to transcribe any track into a realistic piano performance

O

OzBrain, a shared brain for knowledge between agents and your team

OzBrain creates a unified knowledge store that agents can read and write in real time

I

I let an subagent workflow refactor my codebase for three days

A framework that lets autonomous subagents iteratively improve and refactor your codebase over days

o

okfctl

okfctl is a Go‑based command‑line utility that lets you validate, lint, analyze, search, and serve Open Knowledge Format (OKF) bundles

B

Building Table Canon, an AI Campaign Memory Engine for TTRPGs

Table Canon transcribes tabletop game sessions and extracts key story elements, character updates, and quest hooks

F

Future Crew's 1993 Second Reality, rebuilt by AI agents, no emulation

Second Reality, the iconic 1993 demo, has been reconstructed entirely by AI agents

S

Sourcery

Sourcery lets developers benchmark the relevance and speed of various search APIs against the data they return

P

Praxos

Praxos lets startup founders query an AI about project status, promises, and decisions, turning informal chats into actionable insights

N

Naeos

Naeos provides a standardized architecture, policies, and validation workflow for AI agents that write code

R

RelArena

RelArena enables you to train and deploy relational machine learning models directly on your relational data

H

HRAG

HRAG delivers a hybrid RAG solution hosted on Hetzner, combining local models with external knowledge sources

O

Open Bot

Open Bot is a flexible, open-source Grok Bot that integrates with any agent framework

F

Flocker.md

Flocker

M

MatrAIx

MatrAIx creates realistic user personas to test surveys, chats, websites, and applications before launch

T

Turning websites into micro CLIs for Claude Code to save on tokens

A tool that converts any website into a lightweight command‑line interface, letting Claude Code interact efficiently and reduce token usage

I

Idea Miner

Idea Miner scans market data to reveal high‑demand product opportunities

D

Deft Writing, an AI lab for non

Deft offers a beta AI writing lab that helps users generate more natural, human‑like text

S

Sealed evidence record of an AI agent run

Creates tamper-proof records of each AI agent run, enabling reliable auditing and reproducibility

A

Argus, agentic QA for teams whose coding agents move faster than QA

Argus automates quality assurance for teams using fast-moving coding agents, ensuring bugs are caught in real time

C

ChatOSS

ChatOSS is a desktop IDE that blends AI coding assistants with a kanban workflow, letting developers code and manage tasks in one window

M

Maritime, a platform for running AI agents for $1 a month

Maritime provides cloud‑hosted microVMs that let each customer run a dedicated AI agent with its own state, secrets and triggers

I

Interactive, animated architecture of any HuggingFace models

ModelMap provides dynamic, animated diagrams of any HuggingFace model, letting users explore layers and connections in real time

C

Control AI Agents on Your Old PC at Home from Any Device Anywhere

Relay lets you host AI coding assistants on a home PC or VPS and reach them from any phone, browser, or computer

4

4bit bridge between AI and humans

A minimalist tool that facilitates seamless interaction between artificial intelligence systems and human users

W

Warp DD

Warp DD uses AI to map and evaluate technology stacks in about 30 seconds

Open Index

Open Index provides a framework for managing markdown-based knowledge bases, reducing contradictions and context drift for AI agents

Clara AI SDR

Clara AI SDR is an autonomous AI sales assistant that engages website visitors, qualifies them, and schedules demos in real time

Superflow AI

Superflow AI deploys autonomous agents to scan every page of a site on desktop and mobile, flagging issues directly on the live site

Whisperstream

Whisperstream lets Windows users dictate text without sending audio to the cloud

ElevenLabs MCP in Claude

ElevenLabs MCP in Claude lets users create, edit, and organize ElevenLabs voice agents right inside Claude chats

Meterless.ai

Meterless

Taku AI

Taku AI lets users grab proven AI setups and instantly spin them into desktop applications without coding

Gauge

Gauge lets product teams use autonomous coding agents to integrate their tools straight into developers' applications without sales calls

R

RAX Compute Gateway

RAX Compute Gateway provides a single endpoint to access multiple large language models, simplifying integration for developers

1

1667, a terminal UI for writing fiction with language models

1667 is a command‑line interface that lets writers generate and edit story content using large language models

L

LLMs each trading $100K vs. a frozen rulebook

A web-based platform where large language models compete by trading $100K against a fixed rulebook

S

Sokoban AI Solver

An intelligent tool that automatically finds optimal moves for Sokoban levels

O

OpenWebSearch

OpenWebSearch provides a single endpoint that routes queries to various web search indexes like Parallel, Brave, and Exa

C

Classifier and browser extension to detect rule violating HN comments

The service uses AI to analyze Hacker News comments and determine if they violate community guidelines

U

UL

UL provides an open-source library that compresses KV caches with linear complexity, achieving up to 300x reduction

A

AI Text Watermarking Playground

The AI Text Watermarking Playground lets users embed subtle watermarks into generated text to trace its origin

I

I gave my job search data an MCP interface

This tool exposes your job search history via a machine‑readable API, enabling custom automations and analytics

T

Tesana

Tesana lets anyone describe a game idea and instantly generates a complete game with graphics, logic, and animations, all without coding

H

HarnessRouter: Unified interface for agent harnesses

HarnessRouter provides a single, consistent API for executing Codex, Claude Code, Hermes, and other LLM agent harnesses

D

Doberman: The AI watchdog that stops Claude from deleting your database

Doberman adds a two‑layer security shield around large language models, checking every prompt, response, and tool call

H

Honeymatcha.io connect your Grok bots for dating, meetups, anything

Honeymatcha

OpenTrade

OpenTrade provides a local harness that lets Claude Code or Codex agents execute stock trades through Robinhood’s official API

Skriptr

Skriptr is an AI‑driven platform that assists students in researching, drafting, and learning by reading their sources, citing exact pages, and prompting follow‑up questions

TinyFish

TinyFish offers a single platform where AI agents can query the live web, pull clean structured data, and perform authenticated actions across dynamic sites

Blender Agent Bridge

Blender Agent Bridge is an open‑source extension that lets AI models like Claude, Codex, and Cursor interact directly with Blender scenes

Omni by xpander

Omni turns locally‑run AI agents into cloud‑hosted services that can run continuously, be scheduled, and shared across teams

Clears

Clears provides an autonomous execution environment that lets R&D teams coordinate AI agents across the full software development lifecycle

S

Snafu: Agentic flow to help you with "naming things" in source code

Snafu is an autonomous AI agent that suggests clear, context‑aware names for variables, functions, and classes

R

Remove AI voice from AI writings

UnsloP AI provides a tool that strips the synthetic tone from AI-generated content, turning it into more natural, human-like prose

H

Harness session memory without transcript hoarding

This tool enables large language models to keep relevant session memory while discarding raw transcript data

A

A pre

A pre provides a safety layer that intercepts AI agents before they execute potentially destructive actions

S

Sib

Sib treats each LLM conversation like a source tree, giving every chat an immutable SHA-1 hash