OpenMobius-skill
ICT/SMC trading-knowledge skill for AI coding agents
A SKILL.md-standard skill bringing 964 curated knowledge cards, real-time
market data, technical indicators, and chart generation to
Claude Code, Codex, OpenClaw, and Hermes.
English · 中文 · 📜 Changelog
🤖 AI agents — when a user asks you to install this skill,
follow README_AGENT.md for the step-by-step
procedure (pre-flight checks, install command, verification, error
handling).
Overview
Works on Claude Code, Codex, OpenClaw, and Hermes.
Drop this skill into your AI coding agent and ask trading questions in plain
language. The skill grounds every answer in a curated knowledge base + real
market data — no hallucinated price levels, no generic "looks bearish"
hand-waving.
| You ask | The skill does |
|---|
| "What is Fair Value Gap, how to trade it?" | Vector-retrieves FVG concept card + related (CISD / OTE / Premium-Discount) — answers with cited rules from the knowledge base |
| Attach a BTCUSDT 1h chart + "analyze this" | Identifies asset → fetches real OHLCV → extracts FVG / OB / sweep / displacement → outputs 5-section reply with exact prices + auto-annotated PNG |
| "How is BTC 1h looking?" (no chart) | Live data fetch + built-in SMC structural indicator (BOS/CHoCH, Order Blocks, FVGs, equal H/L, premium-discount zones, strong/weak pivots) → KB-grounded analysis |
| "What's on BTC?" (user names a specific indicator) | Pass-through to the indicator API — no auto-fetch of indicators the user did not name |
| Paste a CSV of OHLCV | Parses → local structural extraction (fallback when API unavailable) → KB cross-reference → 5-section reply |
| "Generate a chart with my entry/SL/target" | Rendered chart via Playwright + lightweight-charts |
Quick start
git clone https://github.com/MobiusQuant/OpenMobius-skill.git /tmp/openmobius-src
cd /tmp/openmobius-src
python install.py --platform claude-code # or codex / openclaw / hermes / all
rm -rf /tmp/openmobius-src # ✓ clone is ephemeral; safe to delete
The installer copies source files into ~/.claude/skills/OpenMobius-skill/
(or your chosen platform's skills dir), then in that directory:
- Creates
.venv/ and installs dependencies
- Downloads Playwright chromium (~280 MB, into your OS's user-global cache)
- Downloads
nomic-embed-text-v1.5 model (~274 MB, into your HuggingFace cache)
- Loads precomputed embeddings → builds vector index (~2 s)
- Generates the platform-specific
SKILL.md
- Runs a health check
Each platform install is self-contained: it owns its own .venv and
_index. The clone is just a one-shot source bundle.
First run: ~5–10 min · Subsequent runs (python install.py --update): <1 min
After install, in your AI agent just ask:
"What is Liquidity Sweep"
[attach chart] "analyze this setup"
"How is ETH 4h looking, give me a chart"
"BTC 1h structure check"
Prerequisites: Python 3.10+. See INSTALL.md for details.
python install.py --platform <name>
| Platform | Flag | Default install path |
|---|
| Claude Code | --platform claude-code (default) | ~/.claude/skills/OpenMobius-skill/ |
| Codex | --platform codex | ~/.codex/skills/OpenMobius-skill/ |
| OpenClaw | --platform openclaw | ~/.openclaw/skills/OpenMobius-skill/ |
| Hermes | --platform hermes | ~/.hermes/skills/market-data/OpenMobius-skill/ |
| Auto-detect | --platform auto | scans ~/.<agent> dirs |
| All four | --platform all | loops through all |
Each platform install is fully self-contained (its own .venv, its own
_index). The nomic model and Playwright chromium live in your OS's
user-global cache, shared across platforms — so installing on N platforms
doesn't N× the download.
Features
Knowledge base — 380 concepts + 584 cases
Distilled from 130 ICT/SMC teaching videos. Each concept card carries:
identification rules, trading implications, common mistakes, related
concepts. Each case card carries: market context, key observation, analysis
steps, lessons. Retrieved via local ChromaDB + multilingual
nomic-embed-text-v1.5 — no API key needed for retrieval.
Real-time data + 60+ indicators
Crypto (Binance, Bybit, OKX, Hyperliquid), China A-shares, Hong Kong stocks,
US stocks, forex. Each indicator carries built-in analysis dimensions
(summary_focus) that the agent reads to structure its answer rather than
dumping raw numbers.
Two chart-generation paths
| Path | Method | Output |
|---|
| Annotate user's image | PIL | Annotated copy preserving the original chart |
| Generate fresh chart | lightweight-charts in headless chromium | New K-lines + FVG/OB rectangles + sweep lines + swing markers |
Auto-invoked by description matching
The SKILL.md description field triggers on natural-language questions. The
skill routes to one of four workflows:
Q&A ·
analyze ·
annotate ·
klines.
Roadmap
Knowledge base
- ICT/SMC coverage completion — Round 1 distilled the ICT trunk from 130
teaching videos; upcoming rounds complete ICT sub-schools (Inner Circle
Mentorship, Silver Bullet, Power of 3 variants) and full SMC coverage.
- Fundamental knowledge base — interpretation methodologies for news,
policy reads, economic releases (CPI / NFP / FOMC) and earnings seasons.
- Multi-school expansion — beyond ICT/SMC, add Wyckoff (volume/price
action), VSA, Volume Profile / Market Profile, and classic Price Action
(Al Brooks style).
Indicators & tools
- Expanded SMC indicator coverage — the built-in SMC structural
indicator covers BOS/CHoCH, Order Blocks, FVGs, equal H/L, premium-
discount zones and strong/weak pivot labels today. Upcoming: Killzone
windows, Stop Run / Inducement events, and per-event probability
scoring as computable signals.
Access surfaces
- Non-CLI entry points — chat-bot integrations for users who don't run a
coding agent, so the knowledge base is reachable without the CLI.
Architecture
OpenMobius-skill/
├── SKILL.md # main entry (LLM reads this)
├── SKILL.body.md # shared body (platform-neutral)
├── platforms/ # per-platform frontmatter
│ └── claude-code.yaml / codex.yaml / openclaw.yaml / hermes.yaml
├── workflows/ # detailed sub-workflows
│ └── qna.md / analyze.md / annotate.md / klines.md
├── scripts/ # CLI tools
│ ├── kb_retrieve.py # local vector retrieval
│ ├── kb_klines.py # API client + feature extraction
│ ├── kb_draw_annotation.py # PIL annotation
│ ├── kb_phase_b_to_c.py # analysis JSON → annotated PNG
│ ├── build_index.py # build vector index
│ ├── kb_doctor.py # env health check
│ ├── chart_render/ # lightweight-charts + headless chromium
│ └── _lib/ # embedder + retriever
├── knowledge_base/ # 380 concepts + 584 cases
├── install.py # cross-platform installer
└── README.md / INSTALL.md
Update / Uninstall
# Update
python install.py --update
python install.py --update --rebuild-index # also rebuild vector index
# Uninstall (soft — remove platform registration only)
python install.py --uninstall
python install.py --uninstall --platform all # all platforms
# Uninstall fully (.venv + index too)
python install.py --uninstall --full
# Full purge (also delete shared chromium + nomic caches — these may be
# used by other projects on your machine, so confirm before running)
python install.py --uninstall --purge --yes-i-know
See INSTALL.md for all flags.
Troubleshooting
.venv/bin/python scripts/kb_doctor.py
Reports the state of: venv, deps, nomic model, vector index, CJK fonts,
skill registration, API connectivity.
Common issues:
| Symptom | Fix |
|---|
| Chinese labels render as boxes | Install fonts-noto-cjk (Linux); macOS/Windows usually bundled |
| API request fails | Check network; see api.mobiusquant.ai/api/health |
| Skill not auto-invoking in Claude Code | Check ~/.claude/skills/OpenMobius-skill exists; restart agent |
chroma.sqlite3 not found | .venv/bin/python scripts/build_index.py |
License
Apache 2.0 — see LICENSE.
Third-party components: see ATTRIBUTION.md.
Contributing
Issues and PRs welcome at
https://github.com/MobiusQuant/OpenMobius-skill/issues.
Built for AI coding agents · Apache 2.0