小贝 (xiaobei) is an AI-powered multi-agent content marketing automation platform for small and medium businesses, built on top of the OpenClaw agent framework. It focuses on producing, publishing, and tracking content across major Chinese social media platforms (WeChat, Douyin, Xiaohongshu, Bilibili, etc.). The system uses a custom forked browser automation CLI (camoufox-cli) for platform interaction, a relay service for API signing, and SQLite databases for content and performance metrics. It includes multiple specialized agent crews (main, IT engineer, content producer, sales-cs) that collaborate through a modular skill system with thin wrapper scripts for agent invocation. The codebase demonstrates solid architecture with consistent patterns (DNA style analysis, workflow steps, patch application), but is heavily coupled to specific platform APIs and DOM structures, making it fragile to upstream changes. It is an actively developed, production-grade application aimed at automating social media operations.
Auto-Redbook-Skills is a CLI tool for creating and publishing Xiaohongshu (Little Red Book) social media notes. It takes Markdown content with YAML front matter, renders it into a cover image and multiple card images using Playwright (headless browser), and optionally publishes the note to Xiaohongshu via a built-in, reverse-engineered HTTP signature client. The tool supports 8 visual themes (e.g., sketch, neo-brutalism, botanical) and 4 pagination modes (manual separator, auto-fit, auto-split, dynamic height). The publishing pipeline is entirely self-contained within the repository, using a vendor module that implements the Xiaohongshu creator platform's authentication, signing, and upload protocols without external API services. The project is primarily a content creation automation tool, not a data research project, though it does involve some data transformation (Markdown to images).
Reels-AutoPilot is a Python automation application that scrapes short-form video (Instagram Reels and YouTube Shorts) from configured creator accounts and reposts that content to the user's own Instagram account, including optional story cross-posting. The architecture is a small monolith of loosely coupled scripts sharing a SQLite database: app.py is the main orchestrator running an infinite scheduling loop, reels.py and shorts.py scrape content via instagrapi/yt-dlp, poster.py uploads clips and stories, remover.py purges posted files, and db.py/helpers.py provide SQLAlchemy persistence plus a config system where SQLite values override static defaults in config.py. A Rich-based terminal dashboard (dashboard.py) shows live posting status, and start.py is an interactive CLI setup wizard. The core data flow is: scrape → download to downloads/ → insert metadata row into the 'reels' table as unposted → poster loop picks the oldest unposted row → upload as an Instagram clip → mark posted → optionally trim to 15s and post as a story with mention/link/hashtag/media stickers → remover deletes the local file after a configurable interval. External dependencies are the unofficial Instagram client instagrapi, the YouTube Data API v3 with yt-dlp, moviepy for video trimming, SQLAlchemy, and Rich. The code is functional but rough: reels.py has a filter bug (a tuple instead of a boolean condition) that lets non-reel media into the database; dashboard.py compares string config values to integers so status toggles always render as 'Off'; start.py validates the wrong variable in two input loops; exceptions are silently swallowed; and Instagram credentials are stored in plaintext in both config.py and the SQLite DB, with database/sqlite.db committed to the repository (a credential-leak risk). There are no tests or CI, and the README advertises features (scheduling, multi-account, auto-hashtags, analytics) that do not exist in the code. Dependency versions are somewhat dated, with moviepy 1.0.3 (circa 2020) the most notable lag. The YouTube channel-resolution helpers and the instagrapi story-sticker composition are the most genuinely reusable pieces.