Most Popular
3,528 documents
JSON1 OT Spec
The JSON OT type is designed to allow concurrent edits in an arbitrary JSON document.
Writing Effective Skills
What makes a skill actually work vs. being ignored or misapplied. Based on studying production skills across Claude Code (Superpowers, Trail of Bits, Anthropic's official plugins), Codex (babysit-pr, skill-creator, curated catalog), OpenClaw (55 bundled skills, 13,700+ community), and Cursor/Cline rule systems (BMAD-METHOD, RIPER-5, steipete/agent-rules).
LLM Evaluation — Interview Grill
> 70+ active-recall questions. Pair with `LLM_EVALUATION_DEEP_DIVE.md`.
agentmark — Benchmark AI Coding Agents on Your Codebase
Build an open-source Python CLI that lets developers benchmark and compare
内置 Agent 提示词
Claude Code 内置了 6 个子 Agent，各自有独立的系统提示词和工具权限，用于分工处理不同类型的子任务。以下是每个 Agent 的完整系统提示词。
Prompt Craft Reference
Production-quality prompt engineering for HeyGen Video Agent. Combines official HeyGen guidance with patterns validated across 80+ test videos.
Architecture Ticket Management System
The Ticket Management System is built as a **microservices architecture** with 7 core services, each handling specific business domains. The system is designed for high scalability, fault tolerance, and maintainability.
GREEN NODE - Project Overview
Full-stack recycling platform connecting waste generators with verified collectors in Cochabamba, Bolivia.
Compliance of Implementations with Specification
The following tables show which features are implemented by the Haskell OpenTelemetry
LLM Evaluation & Benchmarking
Generative models produce **open-ended text** — there is rarely a single “correct” string. Quality is **subjective**, **multi-dimensional**, and **context-dependent**: the same answer can be excellent for a casual user and unacceptable for a regulated workflow. Without a disciplined evaluation strategy, teams ship models that look good on a leaderboard but fail in production, leak unsafe content, or hallucinate in high-stakes domains.
CodeForge AI - Hackathon Demo Script
**[Fade in: Dark background with CodeForge AI logo animating in]**
Zero-Touch Training — Developer Summary
**Repo:** [github.com/parkercombes/zero-touch-training](https://github.com/parkercombes/zero-touch-training)
Continue.dev MCP Integration Setup Guide
Edit your Continue.dev configuration file:
PCB Design Sources for Industry-Scale Testing
**Where to find real PCB designs for testing optimization on large, industry-scale boards**
Vibe Marketing Skills v2.1 — Architecture Deep Dive
**Version:** v2.1 | **Date:** March 2026 | **Platforms:** Claude Code, OpenAI Codex, GitHub Copilot CLI
AGENTS.md - Codex-Synaptic Agent System Architecture
The Codex-Synaptic system enhances OpenAI's Codex with advanced multi-agent capabilities, featuring MCP/A2A bridging, neural meshes, swarm coordination, topological constraints, and various consensus mechanisms. This document outlines the comprehensive agent architecture and deployment strategies.
Judge-Tuner: LLM Program Evaluation Suite Builder
Judge-Tuner is an application designed to create and improve evaluation suites for LLM (Large Language Model) programs. It leverages the EvalForge library to enhance its evaluation capabilities.
DunApp PWA - Project Constraints
> **⚠️ KRITIKUS DOKUMENTUM**
Kontext — Context Engine for AI Coding Agents
> Your AI agent is only as good as its context. Stop feeding it garbage.
FreeRangeNotify — Simplification Roadmap
> **Audience**: Product, Engineering, Sales
Contents
<h1 align="center">A Collection of Text-to-Image Generation Studies</h1>
AI_persona
You are a deep-thinking CodeNavigator, an elite AI coding assistant specializing in comprehensive codebase management, systematic debugging, and strategic code improvement. Your core purpose is helping developers maintain and enhance complex codebases with surgical precision and architectural foresight. You may use an extremely long chain of thoughts to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct or most optimal solution b
🤖 Agentic Finance Director — Agent Inventory (Batch 3: AGT-101 → AGT-150)
> **50 Agents | Cross-Cutting, RAG Infrastructure, Data Pipeline, Multi-Agent Orchestration, Advanced Analytics**
Dataset setup
The code for Human3.6M data preparation is borrowed from [VideoPose3D](https://github.com/facebookresearch/VideoPose3D), [SemGCN](https://github.com/garyzhao/SemGCN), [EvoSkeleton](https://github.com/Nicholasli1995/EvoSkeleton).