← Back to list
volatility-modeling Checked
Volatility estimation, forecasting, and regime classification using GARCH, EWMA, realized volatil…
8.0Overall rating
Installation
🤖 Install via AI
Copy the prompt below and send it to your AI assistant (e.g. Claude Code) — it will follow the instructions and install automatically.
Install the "volatility-modeling" skill by following the instructions at https://skill123.me/install/volatility-modeling.
⌨️ Command line install
Run in your terminal — downloads and installs to ~/.claude/skills/.
curl -fsSL https://skill123.me/install/volatility-modeling.sh | bash
📦 Download ZIP
Download the zip and extract it into your skills directory (e.g. ~/.claude/skills/), then restart your session.
⬇ Download v1.0.0 · 20 KBAbout this skill
Overview
Volatility estimation, forecasting, and regime classification using GARCH, EWMA, realized volatility, and volatility cones
Source
- Repo: https://github.com/agiprolabs/claude-trading-skills
- Path: skills/volatility-modeling
Score breakdown
Trigger
4.0
Description is keyword-rich (GARCH, EWMA, realized volatility, cones, regime classification) but lacks When/When-Not and trigger variants: 3+2+3+0=8/20
Structure
9.3
265-line SKILL.md, two references and two scripts in clean tables, inline load cues: 4+5+3+2=14/15
Workflow
7.2
Leans reference/knowledge-style (estimator catalog) rather than step-by-step workflow, but scripts make the math executable; API-key and dependency guards present; light verification guidance: 4+4+5+3+2=18/25
Content
10.0
Precise formulas with parameter guidance, runnable examples: 10/10
Engineering
10.0
Valid frontmatter; shebangs, declared deps incl. optional httpx, env vars documented in script headers; all paths exist: 10/10
Security
10.0
Only declared egress (Birdeye OHLCV with env key, opt-in via --live); no eval/base64/destructive ops, no secrets, no injection; capabilities match declared scope: 20/20
Comprehensive volatility estimation skill: five estimators (close-to-close, Parkinson, Garman-Klass, EWMA, GARCH), cones, regime tables, and two scripts with offline demo plus optional --live Birdeye mode. Capability manifest: reads BIRDEYE_API_KEY/TOKEN_MINT env only with --live; no local file/credential reads; no writes/persistence; sole egress is GET public-api.birdeye.so (declared in script header, consistent with 'optional for live data' dependency note). Capabilities match declared scope.