Stats
Actions
Tags
From mltoolkit
Build an anomaly detection pipeline — IsolationForest, LOF, EllipticEnvelope, OneClassSVM. Ranks anomalies and visualizes on PCA.
How this skill is triggered — by the user, by Claude, or both
Slash command
/mltoolkit:anomalyThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
- Session scratchpad: `.mltoolkit/session.py`
.mltoolkit/session.py{SKILL_DIR}/references/anomaly_reference.py.mltoolkit/:
python {PLUGIN_ROOT}/scripts/stage_session.py --task anomaly --dest .mltoolkit
This copies anomaly_reference.py as session.py, plus model_zoo.py and the _shared/ package, co-located.python .mltoolkit/session.py --data <DATA> --output-dir .mltoolkit --stage compare --contamination 0.05python .mltoolkit/session.py --data <DATA> --output-dir .mltoolkit --stage all --model iforest --contamination 0.05mltoolkit:package..joblib saved) — note this to the user.npx claudepluginhub olatechie/mltoolkit-pluginCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.