tsingyuai/growth-lab

Bilibili Media Crawler collects Bilibili video and creator evidence via search, BV detail, comments, dynamics, and optional media downloads for topic, format, title, creator, or audience research with explicit time-range and quality controls.

Create SEO Page turns validated search demand and SERP evidence into complete SEO pages within a product's existing codebase. Use it when you need to design, write, and implement tool, resource, tutorial, comparison, or solution pages with metadata, structured data, internal links, and engineering verification.

Douyin Media Crawler collects Douyin competitive evidence with MediaCrawler using keyword search, exact video detail, comments, media, and creator profiles. Use it for trend, hook, format, audience-language, or creator research that needs reproducible raw evidence and a documented selection method.

Growth Lab Onboarding audits and configures all Growth Lab dependencies through natural language, checking API keys, third-party CLIs, browsers, and login states, and reports what is ready, missing, optional, or bypassed for first-time setup, environment checks.

Image Generation generates, edits, or batch-creates SEO, marketing, social, and document images using Gemini or OpenAI. Use it when a page or content loop needs illustrations, diagrams, covers, reference-image edits, multiple candidates, or verified image text and composition.

Media Crawler installs, authenticates, configures, operates, and troubleshoots the shared MediaCrawler client used by Douyin, Kuaishou, Bilibili, Weibo, Tieba, and Zhihu collectors. Use it when auditing the client, onboarding a supported platform account, selecting search/detail/creator modes, enabling comments or media, locating outputs, or diagnosing crawler failures.

Media Crawler Kuaishou collects Kuaishou competitive evidence using keyword search, video details, comments, media, and creator profiles. Use for trend, format, audience-language, or creator research requiring reproducible source records and explicit collection limits.

Media Crawler Tieba collects Baidu Tieba threads and user evidence using keyword or bar discovery, exact thread detail, replies, and creator pages. Use it for community pain-point, vocabulary, objection, topic, or user research while preserving thread context.

Progressive Product Research guides an agent to progressively study a product from code, URLs, or user input, recording stable, traceable findings to SOUL.md while keeping facts, hypotheses, and unknowns distinct.

SEO Demand Research identifies real search demand by expanding keyword matrices, validating Bing impressions, and scraping live SERPs, then analyzes top pages across search engine, user, and quality layers to surface information gaps before you create SEO pages.

SEO Page Loop coordinates an observation-action-review loop that researches demand, creates and reviews SEO pages, generates images, submits to IndexNow, and measures performance with persistent memory. Use it to take an SEO page from opportunity discovery through publication and iteration.

SEO Page Review audits SEO landing pages against real search intent, checking content value, AI tone, evidence, metadata, and image relevance. It classifies issues by severity and recommends keep, merge, rewrite, or delete.

SEO Performance Review examines Bing Webmaster page and query data, index status, product results, and Bing AI Performance citations to diagnose search and AI visibility problems. Use it to compare periods, analyze cited pages and grounding queries, and decide the next SEO or GEO action.

Screenshot Assets captures authenticated product screenshots through a Playwright CDP script and archives them in the invoking loop's Memory for truthful UI evidence. Use when you need reusable product screenshots, a specific component, or a full-page browser capture without MCP tools.

Weibo Media Crawler collects Weibo posts and creator evidence using MediaCrawler with search, exact post IDs, comments, optional media, and creator IDs. Use it for discourse, trend, messaging, audience-language, or account research that requires preserved provenance and risk-aware detail enrichment.

XHS Replicate creates Xiaohongshu content from a single viral reference, covering research, structure migration, real product content, image generation, compliance checks, and 24h/48h/7d result review.

Xiaohongshu Card Renderer converts approved Xiaohongshu drafts, analysis references, and product assets into reviewable social cards using deterministic HTML/CSS/SVG/Canvas/Pillow rendering or optional AI image layers. Use it before publishing when you need to lock copy, capture real product evidence, and produce PNG packs with manifest validation and compliance checks.

Xiaohongshu Research searches Xiaohongshu and downloads candidate cover images for visual reference review. Use it to gather read-only evidence for xhs-replicate topic and visual research while saving sanitized evidence locally.

Zhihu Media Crawler collects Zhihu answers, articles, videos, comments, and creator evidence with MediaCrawler through search and exact URLs. Use it for expert discourse, problem framing, objections, terminology, topic, or creator research.