MCP App StoreMCP App Store
from agentic-kaggle-skill53
Assistant skill to run and manage Kaggle competition workflows: submission troubleshooting, score stabilization patterns, kernel workflows, and spec-driven dele
This skill packages practical, battle-tested patterns for running Kaggle competitions with AI agents. It teaches the agent how to set up projects, automate routine monitoring, prepare and push kernels, handle common submission errors, and use spec-driven development to delegate code and experiments to coding agents. The goal is reliable, repeatable competition workflows that reduce manual overhead and improve leaderboard performance.
Use this skill when starting or maintaining a Kaggle competition project, when submissions fail or return unexpected results (400 errors, format issues), when you need to reproduce top notebooks, or when you want to safely delegate GPU training and environment setup to subagents. It’s also useful for stabilizing noisy leaderboard scores by following time-based evaluation guidance.
kaggle CLI commands and kernel management.Best used with coding/delegation-capable agents (OpenCode, Claude Code, or subagent delegates) for tasks like pushing kernels, editing SPEC.md, and orchestrating training runs.
Kaggle competition workflow skill with comprehensive SKILL.md covering end-to-end ML competition strategy and 4 helper scripts. Scripts scaffold project structure, create folds, prepare Kaggle kernel metadata, and prepare dataset metadata. Only scaffold_competition.py runs without required args; the other 3 correctly require input arguments and print usage on missing args. Clean, well-structured code with no security concerns.
Thorough competition workflow skill. No security issues found. Scripts are CLI tools requiring explicit arguments — no dangerous defaults. The experiment_logger embedded in prepare_kaggle_kernel.py runs `git rev-parse HEAD` via subprocess but this is safe/read-only.
Development Worktree
Create an isolated git worktree for feature work, auto-run project setup, and verify a clean test baseline before development.
ds-fix — data-science mid-analysis fixer
Orchestrates diagnosis and targeted fixes mid-analysis: diagnose root cause, apply fixes with output-first verification, and update project learnings.
Readwise Reader Document Management
Manage Readwise Reader documents: list, save, search, move, tag, highlight, export and bulk-edit via official and custom CLIs.
Bounty Hunter — Atlas
Persona skill: 'Atlas' — a profit-focused developer persona for discovering, evaluating and executing paid bounties or freelance tasks with ROI-aware workflows.
Junshi — Research Advisor
Daily strategic research advisor that scans arXiv/venues, digests papers, and proposes bold, ranked research ideas tailored to the user's profile.
Scikit-learn (Classical ML reference)
Comprehensive scikit-learn guidance for classification, regression, clustering, preprocessing, model evaluation, and production-ready ML pipelines.
Full Stack Builder
End-to-end builder that scaffolds, implements, tests, and optionally deploys web and API applications from a natural-language specification.
ezBookkeeping API Tools
Command-line API tools for ezBookkeeping: record and query transactions, retrieve accounts/categories/tags, and fetch exchange rates for self-hosted personal fi