Bridle

Bridle is an open protocol that connects your coding agent to your teammate's.

VTCode

A Rust agent for safe, reviewable coding work in the terminal

Provides tools, context, sandboxing, state management, and verification in one runtime so developers can handle interactive tasks and long-running autonomous coding without an IDE.

Ardent AI

Writes code to work with multiple business data sources and turns it into reusable AI capabilities within a company.

Helps companies adopt custom agents without building development infrastructure themselves, process large datasets from spreadsheets, databases, and APIs in an isolated environment, and reuse the results across the organization.

libpolycall

A C-based runtime broker for connecting programs written in different languages.

A project that helps developers running programs written in multiple languages set up cross-language function calls through a shared runtime and RPC protocol.

AQ

A cloud workspace where teams and multiple coding agents collaborate in one session

Development teams working with multiple AI coding agents can collaborate in real time through a shared terminal, code editor, and app preview, while isolated worktrees for each task help reduce conflicts.

FrontierHarness Eval

An evaluation service comparing quality, cost, and speed across agent harnesses using the same model

Helps developers selecting or validating agent harnesses for software engineering compare success rates, token usage, costs, and latency with the same model and runtime environment.

saccade

An AI agent interface that verifies semantic changes and action results in an approved browser tab

For tasks where automation becomes unreliable after the first click, such as signed-in admin pages or dynamic forms, it helps AI agents keep track of the state of a user-approved Chrome or Edge tab.

rose-1h-nmr

A pretrained model and adaptation protocol for 1H NMR spectrum embeddings.

Provides researchers and developers working with 1H NMR data a pretrained spectrum model and paper based adaptation procedure for inference, search and peak tasks.

Astera

A desktop workbench for coordinating long-running parallel work in Claude Code and Codex

Helps developers running multiple AI coding sessions schedule and resume work, assign dependent tasks to Claude Code and Codex, and run them safely in parallel in separate Git worktrees.

A free collection of web utilities that run in your browser without sending data to a server.

A free service with 16 tools for development, work, and everyday tasks that run quickly in the browser and handle text and image conversion and calculations.

Ramanujan

A distributed computing platform that runs Python scientific calculations on a cluster of idle devices

Researchers and simulation developers can write distributed scientific calculations in Python, compile them for the Ramanujan runtime, and run them using idle computing power from various internet-connected devices.

singular-lite

An orchestration engine for coordinating autonomous AI coding agents in parallel within a repository

A Bash and Python engine for teams running multiple autonomous AI coding agents in a software repository. It coordinates parallel work with 3-level scheduling, persistent leases, an audit pipeline, and Git worktree isolation.

beckon

A tool that signals Claude Code status with distinct sounds and sends terminal notifications over SSH

For Claude Code users who cannot keep watching the terminal, it uses different sounds to signal when an agent finishes, needs a decision, or fails. When running remotely over SSH, it uses the terminal bell and desktop notifications.