OpenMobius-skill equips AI agents with specialized trading frameworks to analyze OHLCV data and charts across crypto, stocks, and forex. It integrates diverse methodologies including ICT/SMC, ChanLun, Wyckoff, and Elliott Wave to provide structured technical analysis.
Quantitative traders can use the skill for chart annotation, price action evaluation, and order flow analysis. The system defaults to strict ICT/SMC logic unless a specific school or data source is requested, ensuring rigorous adherence to selected trading models.
Follow the documented setup, then try a first task.
Installs: Complete self-contained OpenMobius-skill installation for Claude Code at ~/.claude/skills/openmobius-skill/. · Claude Code
Platform: macOS or Linux
In a terminal, verify that Python 3.10 or later is available. Use the macOS/Linux command from this documented block.
python3 --version # macOS / Linux
In a terminal, run this complete block from any directory. It creates a temporary source directory, installs the Claude Code copy, leaves the source directory, and removes only that temporary clone.
Claude
Codex
Cursor
OpenClaw
HermesOPENMOBIUS_SRC="$(mktemp -d "${TMPDIR:-/tmp}/openmobius-src.XXXXXX")"
git clone https://github.com/MobiusQuant/OpenMobius-skill "$OPENMOBIUS_SRC"
cd "$OPENMOBIUS_SRC"
python3 install.py --platform claude-code # → ~/.claude/skills/openmobius-skill/
cd "${TMPDIR:-/tmp}"
rm -rf -- "$OPENMOBIUS_SRC" # ✓ exact mktemp directory onlyIn a terminal, verify the installed copy. Replace the literal `<EXACT_INSTALL_TARGET>` with `~/.claude/skills/openmobius-skill/` and replace `<FLAG>` with `claude-code`; keep both substitutions in the command before running it.
cd <EXACT_INSTALL_TARGET>
.venv/bin/python scripts/kb_doctor.py \
--platform <FLAG> --expected-dir "$PWD"Installs: Local WorkBuddy import of the OpenMobius-skill ZIP built from this repository. The compact package supports School and exact-source lexical retrieval but omits canonical fused-card and vector retrieval. · WorkBuddy
Platform: macOS or Linux terminal plus the WorkBuddy interface
In a terminal, run this complete block to build `/tmp/openmobius-skill-workbuddy.zip` from a temporary clone and remove only that temporary source directory.
OPENMOBIUS_WORKBUDDY_SRC="$(mktemp -d "${TMPDIR:-/tmp}/openmobius-workbuddy.XXXXXX")"
git clone --depth 1 https://github.com/MobiusQuant/OpenMobius-skill "$OPENMOBIUS_WORKBUDDY_SRC"
cd "$OPENMOBIUS_WORKBUDDY_SRC"
python3 scripts/build_workbuddy_package.py \
--output /tmp/openmobius-skill-workbuddy.zip
cd "${TMPDIR:-/tmp}"
rm -rf -- "$OPENMOBIUS_WORKBUDDY_SRC"In WorkBuddy, open Skills → Add Skill → Upload Skill and import `/tmp/openmobius-skill-workbuddy.zip`. Treat installation as complete only after the Skill is visible under Installed.
Uses openmobius-skill
Use openmobius-skill to answer: “What analysis models can I use?” Report the installed native market-analysis profiles, Q&A-only lenses, knowledge categories, and available composition modes without fetching market data.
Uses openmobius-skill
Use openmobius-skill to answer: “What is Liquidity Sweep?” Give a concise knowledge-base-grounded explanation and state any uncertainty rather than fabricating details.
Uses openmobius-skill
Use openmobius-skill to answer: “Compare ICT and ChanLun definitions of market structure.” Keep the two Q&A branches isolated and ground the comparison in attributable knowledge.