-
autonomous-sdlc-harness Firu Daniel (1 project)
An autonomous software-delivery harness: a task prompt becomes a planned, implemented, reviewed and pushed branch, driven by a fleet of single-purpose agents through plan, implement, review, intera...
-
canreuse Marcos Recio (2 projects)
Can I reuse this? A deterministic, directional license-compatibility check: given material under license A and a project under license B, it answers allowed / allowed-with-obligations / review / di...
-
galaxy-brain Marcos Recio (2 projects)
The deterministic harness a coding agent should be standing on: code shape, call graph, diff impact and crash state in milliseconds, offline, with no model in the loop. One CLI, 17 languages, zero ...
-
Gander Arjun Maniyani (1 project)
Gander opens PDF, Word, Excel, PowerPoint, images, video, audio, Markdown and code on Android, entirely on the device. It requests no Android permissions at all and does not hold INTERNET, so it ca...
-
kt-witness GDB Security (1 project)
A transparency log is only a security guarantee if somebody independent is checking it. Key Transparency now ships to billions of users across five production deployments, and until this witness th...
-
Tons of Skills Jeremy Longshore (1 project)
Tons of Skills is an MIT-licensed, model-agnostic platform for discovering, validating, packaging, and installing reusable AI-agent skills. Its harness-free canonical layer, verified adapters, pack...
-
mystic-crypt Asterios Raptis (2 projects)
A cryptography library and command line tool for the JVM: it generates key pairs, converts between the key and certificate formats that tools in this space keep disagreeing about (PEM, DER, PKCS#1,...
-
mystic-crypt-ui Asterios Raptis (2 projects)
The desktop application built on mystic-crypt: a workbench for the keys and secrets a developer or administrator handles every day, with an encrypted local vault as one of its functions rather than...
-
districtry Adam Overberg (1 project)
A dependency-light static web app that answers one question for any point in its coverage: which civic districts contain it, and who represents you there. It reports every boundary at once — ward...
-
NFG Valithor Obsidion (7 projects)
Note Form Generator — a dependency-free HTML/JS tool that turns inline JSON schemas into tabbed forms and exports a standalone page.
-
DiscordPHP-Voice Valithor Obsidion (7 projects)
Voice send and receive for DiscordPHP, including Opus audio and DAVE end-to-end encryption.
-
DiscordPHP Valithor Obsidion (7 projects)
The async PHP library for building Discord bots, built on ReactPHP. Full gateway and REST coverage, an event-driven model, and a large ecosystem of downstream bots and libraries.
-
TwitchPHP-Http Valithor Obsidion (7 projects)
The standalone Helix transport for TwitchPHP — async request queue, points rate-limiting, and typed errors.
-
TwitchPHP Valithor Obsidion (7 projects)
Async Twitch framework for PHP, built like DiscordPHP: Helix REST, EventSub over WebSocket, and IRC chat.
-
DiscordPHP-Http Valithor Obsidion (7 projects)
The standalone async HTTP and rate-limit layer DiscordPHP talks to Discord through — request queueing, bucket rate-limiting, and typed errors. Reusable outside DiscordPHP.
-
DiscordPHP-EventLogger Valithor Obsidion (7 projects)
Drop-in audit logging for DiscordPHP bots — every gateway event, formatted and routed.
-
TrenTorch-Web TrenTorch (2 projects)
The web version of TrenTorch: the same from-scratch, module-by-module ML curriculum, running in the browser, no local install.
-
Learn PyTorch and frontier ML by building your own PyTorch, from scratch, tensor to transformer, just NumPy underneath. 20 progressive modules across classical ML, deep learning foundations, langua...
-
Zeitlos Machdyne UG (haftungsbeschränkt) (1 project)
Zeitlos is an open-source RISC-V SoC and operating system for FPGA computers. It is designed as a complete, user-controlled computer platform, with open FPGA gateware, CPU, operating system, and ap...
-
Quillan-Ronin CrashOverrideX (1 project)
Quillan-Ronin is an open-source sovereign AI runtime: a 34-expert council Mixture-of-Experts architecture with BitNet 1.58-bit quantization, diffusion-based reasoning refinement, and a 9B virtual-a...