Explore
16,091 skills indexed with the new KISS metadata standard.
Like in the original symbiants using a scalar signal field, I want to introduce predation. I'd like it to be a little more nuanced then the way I did it before, just spawnings new factions and weighted randomly selecting a food source based on total number of agents belonging to [each](../docs/Pools
This document collects ideas and instructions for implementing future improvements. Follow these when adding features or refactoring the code.
- `npm run dev` - Build in development mode with watch
Manage Liongard agents with bulk operations and installer generation.
generic skill
This file contains essential information for agentic coding assistants working in this repository.
generic skill
Learn how to interact with Mastra AI agents, including generating responses, streaming interactions, and managing agent tools using the client-js SDK.
CLI forensic tool for Telegram Desktop artifact analysis. Extracts and analyzes data from `tdata` directory.
- **Indentation**: 4 spaces
- `app/` — Next.js App Router routes, layouts, and API handlers (`app/api/**`).
TODO: Add service description.
- [Remix](https://remix.run)
- `src/pages/` 负责声明 Astro 路由，只保留轻量页面逻辑并组合 `layouts/` 与 `components/`。
These instructions define how GitHub Copilot should assist with this Go project. The goal is to ensure consistent, high-quality code generation aligned with Go idioms, the chosen architecture, and our team's best practices.
`docker compose up` to run the docker container
This document outlines key information needed by coding agents to work in this repository. It outlines the rules to follow during work, and provides helpful commands and tips to facilitate development.
This repository contains `mnvkd`, a C server-authoring toolkit built around a virtual kernel with coroutine, actor, service and application frameworks.
This file defines agents and their respective roles for the `scjson` project. Each agent is responsible for a specific transformation, validation, or extraction task.
We are a team of Pokémon trainers utilizing PokeAPI (https://pokeapi.co/) as the primary source
This project generates structured ground truth data for validating calculations used in machine learning framework by using PyTorch for executing and the gradienttracer framework. The data is stored in GGUF format for use in neural network testing and validation.
- `pages/` drives Nuxt 4 routing, while shared UI lives in `components/`; drop reusable logic in `libs/` and server handlers under `server/api/`.
LoRA Prep is a macOS-first toolkit that prepares LoRA training images. It ships both a SwiftPM CLI executable (`LoRAPrep`) and a SwiftUI desktop app; both surfaces share the same Vision/Core Image pipeline and should remain feature-parity replacements for the legacy `loraPrep.sh` script (which now s
`naeural_core/` houses the runtime: `main/` orchestrates startup, `business/` contains pipeline plugins and the testing harness, `comm/` handles transports, and `serving/` wraps model execution. Runtime artifacts land in `_local_cache/`, created on demand. Optional packages live under `extensions/`,