Organize Kaggle ML experiment documentation into a 3-layer structure.
npx skills add https://github.com/daikichiba9511/dotfiles --skill ml-docs
Please help me install this Agent Skill. Skill: ml-docs Source: https://github.com/daikichiba9511/dotfiles/tree/main/private_dot_codex/skills/ml-docs Command: npx skills add https://github.com/daikichiba9511/dotfiles --skill ml-docs
Manage and standardize Kaggle ML experiment documentation by organizing a 3-layer structure: Full Log, Evidence Ledger, and Compact Document, enabling clear history, verified facts, and ready-to-resume summaries.
Run ml-docs init exp001 to create the 3-layer documentation scaffold for a new experiment.
High-intent search queries and answers about installing and using this skill.
Organizing machine learning experiment documentation into a 3-layer structure—Full Log, Evidence Ledger, and Compact Document—separates history, verified facts, and operational context to ensure repeatable experimentation.
Structuring Kaggle competition strategy and results logs involves scaffolding dedicated files for each documentation layer, allowing you to form proposals, record results, and review the current state using lifecycle commands.
Creating an agent-ready summary to resume ML work quickly requires maintaining a Compact Document layer, which provides an operational context summary updated alongside the Full Log and Evidence Ledger.
Yes, you can manage multiple Kaggle project documentations simultaneously by initializing dedicated 3-layer documentation scaffolds for each experiment, enforcing clear update rules across all active projects.
Standardizing ML experiment documentation requires no external dependencies, utilizing built-in lifecycle commands to initialize experiments, formulate strategies, log results, and review ongoing documentation status.
datamonsterrBootstrap a reproducible autoresearch experiment folder with templates and registry updates.
brennacodesGenerate a tailored resume from logged evidence and job listings.
discretedsAutomate daily competition data collection and issue management via cron.
THUFanZdTurn EXPERIMENT_PLAN.md into runnable GPU experiments and collect results.
