MCP App StoreMCP App Store
from claude-ml-intern-skill54
Autonomous ML intern that researches, implements, trains, verifies, and publishes ML experiments using the Hugging Face ecosystem, with budgeted retry loops and
ml-intern is a comprehensive orchestrator skill that guides an agent through the full ML experiment lifecycle using Hugging Face tools. It covers task clarification, research, planning, parallel implementation of solution paths via subagents, smoke-testing, training, self-verification, failure postmortems, and publishing successful runs to the Hugging Face Hub. It enforces reproducibility by writing TASK/PLAN/RESEARCH/RESULTS/VERIFY artifacts and publishing model bundles when runs pass verification.
Use this skill when a user asks the agent to implement, train, fine-tune, or reproduce an ML model or paper (e.g., "fine-tune Qwen on dataset X", "implement DeepSeek-V3 at 100M parameters", or "reproduce paper Y"). It's intended for tasks that require experiment orchestration, reproducible outputs, and automated notifications (Telegram/Slack). It's also appropriate for budgeted automated runs where retries and parallel solution paths are needed.
Designed for advanced agent environments that support spawning subagents, running shell commands, and interacting with the web (Claude Code, agents with subagent/task orchestration). Works best when agents can run CLIs, write files, and access HF tokens for publishing.
This skill has not been reviewed by our automated audit pipeline yet.
Development Worktree
Create an isolated git worktree for feature work, auto-run project setup, and verify a clean test baseline before development.
WRDS Query & ETL Enforcement
Standards and enforcement guidance for querying WRDS data and running SAS/ETL on the WRDS grid—includes query validation, SGE submission patterns, and performan
Academic Research Search
Search academic literature across multiple sources, deduplicate results, resolve DOIs, and surface trusted papers with concise takeaways.
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.
Full Stack Builder
End-to-end builder that scaffolds, implements, tests, and optionally deploys web and API applications from a natural-language specification.