An agentic growth tool executing SEO, Xiaohongshu, and WeChat campaigns using AI coding runtimes.
Run closed-loop marketing workflows using Claude Code or Codex to research, execute, and iterate.
Growth Lab is an open-source agentic growth system that operates directly inside Claude Code or Codex runtimes. It handles product analysis, keyword search, content creation, and post-campaign review across platforms like Xiaohongshu, WeChat, and Web SEO. You should use it if you want an AI workflow that retains long-term memory and updates marketing strategies based on real campaign performance.
The tool turns marketing into a closed loop: observe, act, and review. It reads code repositories or product documentation, generates SEO-optimized pages or platform posts, manages image generation via OpenAI or Gemini API, and feeds performance metrics back into workspace memory files.
It suits growth teams, founders, and technical marketers who want to execute repeatable marketing loops directly from local coding runtimes. It is not suitable for non-technical users looking for a traditional no-code SaaS dashboard with point-and-click menus.
Persists analytics, action results, and context across sessions in local filesystem memory files.
Automates SEO page research, Xiaohongshu note reproduction, and WeChat draft synchronization.
Uses Claude Code or Codex as an application runtime to parse prompts and invoke local skills.
Integrates browser automation tools and official APIs to gather high-performing posts and search trends.
Growth Lab is entirely free and open-source under the Apache-2.0 license. The only costs are your self-hosted infrastructure, API calls for LLM image and text generation, and access to Claude Code or Codex runtimes.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
Growth Lab uses Codex or Claude Code as its execution runtime to read repository files, execute local scripts, and retain persistent memory.
Xiaohongshu publishing requires manual human action, while WeChat articles sync as drafts and only publish under explicit command and triple-confirmation.
All product knowledge, research materials, generated assets, and iteration memories are stored locally in markdown and workspace files.
See Zekai first in Google