MCP App StoreMCP App Store
from zorai308
Integrate Weights & Biases for ML experiment tracking: log metrics, hyperparameters, checkpoints, run sweeps and view collaborative dashboards.
Weights & Biases (wandb) integration lets an agent or developer instrument training runs to log metrics, hyperparameters, model checkpoints, artifacts, and visualizations. The skill shows how to initialize wandb, log training/validation metrics per epoch, finish runs, and run hyperparameter sweeps (Bayesian/agent-driven). It includes example Python snippets for init, wandb.log usage, and an example sweep configuration.
Use this skill when you need reproducible experiment tracking, collaborative dashboards for ML teams, automated hyperparameter search, or versioned model/artifact storage. It's appropriate during model development, hyperparameter tuning, and when you want quick visualizations of training behaviour.
Agents and environments that run Python and can execute training code (Copilot/Codex-style agents, local Python runners, or any assistant that can provide code and instructions).
Documentation-only skill for Weights & Biases ML experiment tracking. Provides clear installation steps, experiment logging, and hyperparameter sweep examples. No bundled scripts to execute. SKILL.md is well-structured with proper frontmatter and idiomatic Python code. Clean security profile with no risks identified.
Pure documentation skill — no scripts to audit for security or runtime behavior. Useful as a reference for wandb integration but lacks automation scripts that would make it more actionable.
Code Optimizer (Performance Audit)
Conducts deep performance audits across databases, memory, algorithms, concurrency, I/O, bundling, and more using specialist agents and pattern-based detection.
Knowledge Base (Help Center)
Guidance and templates for designing help center architecture, writing effective support articles, and optimizing search to maximize self-service deflection.
EconML (Microsoft) — Heterogeneous Treatment Effects
Guidance and examples for using Microsoft EconML to estimate heterogeneous treatment effects (Double ML, Causal Forest, Deep IV) from observational data.
Git Guardrails (Claude Code)
PreToolUse hook that blocks destructive git commands (push, reset --hard, clean, branch -D) so agents like Claude Code cannot run them without approval.
DSPy — Declarative LM Programming
Use DSPy to build declarative, modular LM pipelines, optimize prompts automatically, and assemble reliable RAG/agent systems with structured signatures and opti
Synalinks Framework
Keras-inspired framework for building structured, neuro-symbolic LLM programs with DataModel schemas, modular Programs, and training/optimization tools.
Terradev — Cross‑Cloud GPU Provisioning
Provision and manage GPUs across clouds, create GPU Kubernetes clusters, deploy inference endpoints, and burst local compute to cloud with BYOAPI credential saf
Runtime Communication (research_mvp)
Rules and workflows for messaging, delegation, and task coordination in the research_mvp local multi-agent runtime (leader, researcher, trainer).
Deliberate Practice
Guided framework for accelerating skill acquisition using focused practice, immediate feedback, and progressive challenge—useful for learning technical skills,
MessyData
Generate realistic synthetic datasets with configurable anomalies (missing values, duplicates, invalid categories, bad dates, outliers) using YAML configs, CLI,
vLLM-Omni Video Generation
Generate videos (text→video, image→video, text+image→video) using vLLM-Omni and Wan2.2-style diffusion models, with guidance on parameters and performance trade
Model Deployment (SageMaker & Bedrock)
Generates deployment code and a Jupyter notebook to deploy LoRA fine-tuned Nova or OSS models (from SageMaker Serverless Model Customization) to SageMaker endpo