Most Popular
3,528 documents
Windsurf功能更新
- 可将浏览器上下文共享给 Windsurf
03 — Evaluation Techniques Deep Dive: LLM-as-Judge, Human Evaluation & Retrieval Metrics
> **Interview Reality:** _"What evaluation techniques do you use and why?"_
starProject
- **Description:** Multilingual Document Layout Parsing in a Single Vision-Language Model
Evaluation Harness (Offline + Online)
- Without a harness, you **can't compare** prompts, models, retrieval configs, or costs.
C13.6: DAG Visualization & Workflow Security
> **Parent:** [C13 Monitoring, Logging & Anomaly Detection](C13-Monitoring-and-Logging)
RWKV-Evals
RWKV Evaluation Data presents RWKV's performance across various large language model benchmarks, including Uncheatable Eval, MMLU, RULER, and LongBench.
msitarzewski/agency-agents
date: 2026-03-15T13:27:54+08:00
LLM Judge — Setup & Operations
The **LLM Judge** is LLMTrace's third security detector alongside the
index
title: "Presidio: Open-Source Framework for PII Detection, Redaction & Anonymization"
角色:金牌面试者
<p align="center"><h1>🧠 Awesome Claude Prompts </h1></p>
Dota 2 Analysis Persona
You are a senior Dota 2 analyst providing written analysis of match replays. Your background includes professional play experience (TI qualifiers, 11k peak MMR) and extensive coaching. You deliver insights through structured analysis, not live dialogue.
ArbitragePro Configuration Guide: Complete Setup and Deployment
1. [Project Overview & Architecture](#project-overview--architecture)
Product Requirements Document: B2B SaaS Landing Page
**Product:** AI-Powered Sales Intelligence Platform
AI Agent Builder & Orchestrator
**Purpose:** Conversational AI system that understands user goals, discovers the best tools/frameworks, and automatically builds complete agent solutions with UI, backend, MCP integration, and documentation.
How to Add Resources to Your FastMCP Server
In the Model Context Protocol (MCP), there are three main capabilities:
Skill: Google Stitch for Content Visuals (MCP-Integrated)
name: google-stitch-content
第 4 章:MCP 协议详解
> **MCP 是 Agent 工具的"USB 接口"——统一了工具的发现、调用和授权,但它不解决工具本身的质量问题,也不能让一个烂工具变好用。**
语义块切分 Semantic Chunking

llm-hallucination-survey

Cursor vs Windsurf: Complete Comparison (2025)
[← Back to Main](../README.md)
Trader ROI Playbook (Codex + CI)
Purpose: increase engineering output per hour while keeping quality stable or better.
midjourney-expert
name: midjourney-expert
FacelessForge - AI Content Generator Overview
FacelessForge is a SaaS platform for automated AI-powered video content generation. It enables users to create faceless videos for social media platforms (primarily TikTok) by orchestrating multiple AI services for script generation, voice synthesis, video clip creation, and final assembly.
write-script
Write a full video script for @SketchySurvival101 following all rules in CLAUDE.md.