<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> OpenMobius-skill enhances AI coding agents (Claude Code, Codex, OpenClaw, Hermes) by integrating a comprehensive ICT/SMC trading knowledge base and real-time market data. It provides AI agents with grounded, accurate trading analysis, preventing hallucinations and enabling sophisticated chart interpretation and generation.

How It Works

The skill leverages a local vector index built with nomic-embed-text-v1.5 on 964 curated ICT/SMC knowledge cards, enabling efficient retrieval without API keys. It interfaces with real-time market data feeds for over 60 indicators across various asset classes. Analysis workflows include natural language Q&A, automated chart analysis with structural indicator identification (BOS/CHoCH, OB, FVG), and chart generation/annotation.

Quick Start & Requirements

Installation involves cloning the repository and running python install.py --platform <platform_name> (e.g., claude-code). Prerequisites include Python 3.10+. Setup requires downloading Playwright chromium (~280 MB) and the nomic-embed-text-v1.5 model (~274 MB). Initial setup takes 5-10 minutes; updates are under 1 minute. Official documentation is available via INSTALL.md.

Highlighted Details

• Knowledge Base: 964 cards (380 concepts, 584 cases) derived from 130 ICT/SMC teaching videos.
• Real-time Data: Supports 60+ indicators across Crypto, China A-shares, Hong Kong stocks, US stocks, and forex.
• Chart Generation: Offers annotation of user-provided images or generation of new charts with lightweight-charts, including auto-annotation of key SMC structures.
• SMC Indicator Suite: Built-in analysis covers BOS/CHoCH, Order Blocks, Fair Value Gaps (FVG), equal highs/lows, premium/discount zones, and pivot labels.

Maintenance & Community

The project roadmap includes expanding ICT/SMC coverage, incorporating fundamental analysis interpretation, and adding other trading methodologies (Wyckoff, VSA, Market Profile). Non-CLI access points are planned. Contributions are welcomed via GitHub issues.

Licensing & Compatibility

Licensed under Apache 2.0, permitting broad commercial use and integration. Compatible with specified AI coding agent platforms.

Limitations & Caveats

The initial focus is ICT/SMC; broader trading methodology support is planned. Chart generation relies on external dependencies like Playwright. Potential CJK font rendering issues and API connectivity problems are noted in troubleshooting.