[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-78063":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":21,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":15,"starSnapshotCount":15,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},78063,"OpenMobius-skill","MobiusQuant\u002FOpenMobius-skill","MobiusQuant","ICT\u002FSMC trading-knowledge skill for AI coding agents (Claude Code \u002F Codex \u002F OpenClaw \u002F Hermes)","https:\u002F\u002Fwww.mobiusquant.ai\u002F",null,"Python",385,45,16,0,4,12,358,15,65.99,false,"main",[],"2026-06-12 04:01:23","\u003Cdiv align=\"center\">\n\n# OpenMobius-skill\n\n**ICT\u002FSMC trading-knowledge skill for AI coding agents**\n\nA `SKILL.md`-standard skill bringing 964 curated knowledge cards, real-time\nmarket data, technical indicators, and chart generation to\n**Claude Code**, **Codex**, **OpenClaw**, and **Hermes**.\n\n\u003Cbr>\n\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-Apache_2.0-blue.svg)](.\u002FLICENSE)\n[![Python](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.10%2B-3776ab.svg)](https:\u002F\u002Fwww.python.org\u002F)\n[![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F🌐_website-mobiusquant.ai-FF7E29.svg)](https:\u002F\u002Fwww.mobiusquant.ai\u002F)\n[![Platforms](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fplatforms-Claude_Code_·_Codex_·_OpenClaw_·_Hermes-9333ea.svg)](#platform-support)\n[![Knowledge Base](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fknowledge_base-964_cards-22c55e.svg)](#features)\n\n[English](.\u002FREADME.md) · [中文](.\u002FREADME.zh.md)\n\n\u003C\u002Fdiv>\n\n---\n\n> **🤖 AI agents** — when a user asks you to install this skill,\n> follow [`README_AGENT.md`](.\u002FREADME_AGENT.md) for the step-by-step\n> procedure (pre-flight checks, install command, verification, error\n> handling).\n\n---\n\n## Overview\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\".\u002Fdocs\u002Fassets\u002Fdemo.gif\" alt=\"OpenMobius-skill demo\" width=\"780\">\n  \u003Cbr>\n  \u003Csub>Works on \u003Cb>Claude Code\u003C\u002Fb>, \u003Cb>Codex\u003C\u002Fb>, \u003Cb>OpenClaw\u003C\u002Fb>, and \u003Cb>Hermes\u003C\u002Fb>.\u003C\u002Fsub>\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\nDrop this skill into your AI coding agent and ask trading questions in plain\nlanguage. The skill grounds every answer in a curated knowledge base + real\nmarket data — no hallucinated price levels, no generic \"looks bearish\"\nhand-waving.\n\n| You ask | The skill does |\n|---|---|\n| *\"What is Fair Value Gap, how to trade it?\"* | Vector-retrieves FVG concept card + related (CISD \u002F OTE \u002F Premium-Discount) — answers with cited rules from the knowledge base |\n| *Attach a BTCUSDT 1h chart + \"analyze this\"* | Identifies asset → fetches real OHLCV → extracts FVG \u002F OB \u002F sweep \u002F displacement → outputs 5-section reply with **exact prices** + auto-annotated PNG |\n| *\"How is BTC 1h looking?\"* (no chart) | Live data fetch → feature extraction → KB-grounded analysis |\n| *\"What's RSI(14) and MACD on BTC?\"* | Calls indicator endpoint for values + per-indicator analysis dimensions |\n| *Paste a CSV of OHLCV* | Parses → analyzes → KB cross-reference → 5-section reply |\n| *\"Generate a chart with my entry\u002FSL\u002Ftarget\"* | Rendered chart via Playwright + lightweight-charts |\n\n---\n\n## Quick start\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FMobiusQuant\u002FOpenMobius-skill.git \u002Ftmp\u002Fopenmobius-src\ncd \u002Ftmp\u002Fopenmobius-src\npython install.py --platform claude-code      # or codex \u002F openclaw \u002F hermes \u002F all\n\nrm -rf \u002Ftmp\u002Fopenmobius-src                     # ✓ clone is ephemeral; safe to delete\n```\n\nThe installer copies source files into `~\u002F.claude\u002Fskills\u002FOpenMobius-skill\u002F`\n(or your chosen platform's skills dir), then in that directory:\n\n1. Creates `.venv\u002F` and installs dependencies\n2. Downloads Playwright chromium (~280 MB, into your OS's user-global cache)\n3. Downloads `nomic-embed-text-v1.5` model (~274 MB, into your HuggingFace cache)\n4. Loads precomputed embeddings → builds vector index (~2 s)\n5. Generates the platform-specific `SKILL.md`\n6. Runs a health check\n\nEach platform install is **self-contained**: it owns its own `.venv` and\n`_index`. The clone is just a one-shot source bundle.\n\n**First run**: ~5–10 min · **Subsequent runs** (`python install.py --update`): \u003C1 min\n\nAfter install, in your AI agent just ask:\n\n```\n\"What is Liquidity Sweep\"\n[attach chart] \"analyze this setup\"\n\"How is ETH 4h looking, give me a chart\"\n\"BTC 1h RSI(14) and MACD?\"\n```\n\n> **Prerequisites**: Python 3.10+. See [INSTALL.md](.\u002FINSTALL.md) for details.\n\n---\n\n## Platform support\n\n```bash\npython install.py --platform \u003Cname>\n```\n\n\u003Cdiv align=\"center\">\n\n| Platform | Flag | Default install path |\n|:---|:---|:---|\n| **Claude Code** | `--platform claude-code` *(default)* | `~\u002F.claude\u002Fskills\u002FOpenMobius-skill\u002F` |\n| **Codex** | `--platform codex` | `~\u002F.codex\u002Fskills\u002FOpenMobius-skill\u002F` |\n| **OpenClaw** | `--platform openclaw` | `~\u002F.openclaw\u002Fskills\u002FOpenMobius-skill\u002F` |\n| **Hermes** | `--platform hermes` | `~\u002F.hermes\u002Fskills\u002Fmarket-data\u002FOpenMobius-skill\u002F` |\n| Auto-detect | `--platform auto` | scans `~\u002F.\u003Cagent>` dirs |\n| All four | `--platform all` | loops through all |\n\n\u003C\u002Fdiv>\n\nEach platform install is fully **self-contained** (its own `.venv`, its own\n`_index`). The nomic model and Playwright chromium live in your OS's\nuser-global cache, shared across platforms — so installing on N platforms\ndoesn't N× the download.\n\n---\n\n## Features\n\n### Knowledge base — 380 concepts + 584 cases\n\nDistilled from 130 ICT\u002FSMC teaching videos. Each concept card carries:\nidentification rules, trading implications, common mistakes, related\nconcepts. Each case card carries: market context, key observation, analysis\nsteps, lessons. Retrieved via local ChromaDB + multilingual\n`nomic-embed-text-v1.5` — no API key needed for retrieval.\n\n### Real-time data + 60+ indicators\n\nCrypto (Binance, Bybit, OKX, Hyperliquid), China A-shares, Hong Kong stocks,\nUS stocks, forex. Each indicator carries built-in analysis dimensions\n(`summary_focus`) that the agent reads to structure its answer rather than\ndumping raw numbers.\n\n### Two chart-generation paths\n\n| Path | Method | Output |\n|---|---|---|\n| Annotate user's image | PIL | Annotated copy preserving the original chart |\n| Generate fresh chart | lightweight-charts in headless chromium | New K-lines + FVG\u002FOB rectangles + sweep lines + swing markers |\n\n### Auto-invoked by description matching\n\nThe `SKILL.md` description field triggers on natural-language questions. The\nskill routes to one of four workflows:\n[Q&A](.\u002Fworkflows\u002Fqna.md) ·\n[analyze](.\u002Fworkflows\u002Fanalyze.md) ·\n[annotate](.\u002Fworkflows\u002Fannotate.md) ·\n[klines](.\u002Fworkflows\u002Fklines.md).\n\n---\n\n## Roadmap\n\n**Knowledge base**\n\n- **ICT\u002FSMC coverage completion** — Round 1 distilled the ICT trunk from 130\n  teaching videos; upcoming rounds complete ICT sub-schools (Inner Circle\n  Mentorship, Silver Bullet, Power of 3 variants) and full SMC coverage.\n- **Fundamental knowledge base** — interpretation methodologies for news,\n  policy reads, economic releases (CPI \u002F NFP \u002F FOMC) and earnings seasons.\n- **Multi-school expansion** — beyond ICT\u002FSMC, add Wyckoff (volume\u002Fprice\n  action), VSA, Volume Profile \u002F Market Profile, and classic Price Action\n  (Al Brooks style).\n\n**Indicators & tools**\n\n- **SMC concepts as indicators** — abstract Liquidity Sweep \u002F FVG \u002F Order\n  Block \u002F Killzone into computable indicators, queryable like RSI \u002F MACD,\n  with numeric readings + occurrence frequency.\n\n**Access surfaces**\n\n- **Non-CLI entry points** — chat-bot integrations for users who don't run a\n  coding agent, so the knowledge base is reachable without the CLI.\n\n---\n\n## Architecture\n\n```\nOpenMobius-skill\u002F\n├── SKILL.md                          # main entry (LLM reads this)\n├── SKILL.body.md                     # shared body (platform-neutral)\n├── platforms\u002F                        # per-platform frontmatter\n│   └── claude-code.yaml \u002F codex.yaml \u002F openclaw.yaml \u002F hermes.yaml\n├── workflows\u002F                        # detailed sub-workflows\n│   └── qna.md \u002F analyze.md \u002F annotate.md \u002F klines.md\n├── scripts\u002F                          # CLI tools\n│   ├── kb_retrieve.py                # local vector retrieval\n│   ├── kb_klines.py                  # API client + feature extraction\n│   ├── kb_draw_annotation.py         # PIL annotation\n│   ├── kb_phase_b_to_c.py            # analysis JSON → annotated PNG\n│   ├── build_index.py                # build vector index\n│   ├── kb_doctor.py                  # env health check\n│   ├── chart_render\u002F                 # lightweight-charts + headless chromium\n│   └── _lib\u002F                         # embedder + retriever\n├── knowledge_base\u002F                   # 380 concepts + 584 cases\n├── install.py                        # cross-platform installer\n└── README.md \u002F INSTALL.md\n```\n\n---\n\n## Update \u002F Uninstall\n\n```bash\n# Update\npython install.py --update\npython install.py --update --rebuild-index    # also rebuild vector index\n\n# Uninstall (soft — remove platform registration only)\npython install.py --uninstall\npython install.py --uninstall --platform all  # all platforms\n\n# Uninstall fully (.venv + index too)\npython install.py --uninstall --full\n\n# Full purge (also delete shared chromium + nomic caches — these may be\n# used by other projects on your machine, so confirm before running)\npython install.py --uninstall --purge --yes-i-know\n```\n\nSee [INSTALL.md](.\u002FINSTALL.md) for all flags.\n\n---\n\n## Troubleshooting\n\n```bash\n.venv\u002Fbin\u002Fpython scripts\u002Fkb_doctor.py\n```\n\nReports the state of: venv, deps, nomic model, vector index, CJK fonts,\nskill registration, API connectivity.\n\nCommon issues:\n\n| Symptom | Fix |\n|---|---|\n| Chinese labels render as boxes | Install `fonts-noto-cjk` (Linux); macOS\u002FWindows usually bundled |\n| API request fails | Check network; see `api.mobiusquant.ai\u002Fapi\u002Fhealth` |\n| Skill not auto-invoking in Claude Code | Check `~\u002F.claude\u002Fskills\u002FOpenMobius-skill` exists; restart agent |\n| `chroma.sqlite3` not found | `.venv\u002Fbin\u002Fpython scripts\u002Fbuild_index.py` |\n\n---\n\n## License\n\nApache 2.0 — see [LICENSE](.\u002FLICENSE).\nThird-party components: see [ATTRIBUTION.md](.\u002FATTRIBUTION.md).\n\n## Contributing\n\nIssues and PRs welcome at\n\u003Chttps:\u002F\u002Fgithub.com\u002FMobiusQuant\u002FOpenMobius-skill\u002Fissues>.\n\n\u003Cdiv align=\"center\">\n\u003Csub>Built for AI coding agents · Apache 2.0\u003C\u002Fsub>\n\u003C\u002Fdiv>\n","OpenMobius-skill 是一个专为AI编码代理（如Claude Code、Codex、OpenClaw和Hermes）设计的ICT\u002FSMC交易知识技能。它提供了964张精选的知识卡片，实时市场数据，技术指标计算及图表生成等功能，使用户能够以自然语言形式询问与交易相关的问题，并获得基于实际数据和知识库支持的精确答案。该技能采用Python 3.10+编写，支持多种平台，并通过调用外部API来获取最新的市场信息和技术分析结果。适用于需要进行金融数据分析、策略制定或教育目的的专业人士以及对量化交易感兴趣的开发者。",2,"2026-06-11 03:56:25","CREATED_QUERY"]