MCP App StoreMCP App Store
from tonone10
Build a reproducible ML training-to-serving pipeline: data validation, feature engineering, training, evaluation, and a serving endpoint.
Cortex provides step-by-step guidance to build a production-ready ML pipeline. It walks an engineer through environment detection, data validation, feature engineering, training scripts, evaluation, and serving endpoints, with emphasis on reproducibility and monitoring.
Use Cortex when you need to deliver a complete ML workflow: from raw data to a deployed model. Ideal for classification or regression projects where you need reliable baselines, experiment tracking, and production-grade serving. Trigger on requests like "build ML model", "train a model", or "create prediction pipeline".
Best suited for code-capable agents: Claude Code, Copilot/Codex-style coding assistants, and other agents that can run or author training and serving scripts.
Cortex Model is an ML pipeline builder skill that guides an agent through building a reproducible training-to-serving pipeline covering data validation, feature engineering, training, evaluation, and serving. No bundled scripts were present for runtime testing. The SKILL.md provides solid best-practice guidance (baseline-first approach, training/serving skew awareness) but lacks concrete code templates and references an external docs/output-kit.md not included in the skill.
Clean skill with good ML engineering principles. Architecture could improve by adding scripts/ with templates or references/ with the output-kit doc. The /atlas-report invocation is unclear — may be a team-internal command not available to other users.
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
Harness Health Audit
Assess and score the overall health of a Claude Code harness across architecture, skills coverage, hooks, rules, MCP servers, eval pipelines, and team setup.
LLM Evaluation
Evaluation framework and tools for systematically measuring LLM performance using automated metrics, human judgment, and A/B testing.
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
Security Hygiene
Security hygiene guidelines for self-modifying agent systems: path sanitization, safe YAML handling, data poisoning checks, and staging/quarantine practices to
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
Code Evaluation Harness
Spawns an independent evaluator agent to score code outputs on functionality, code quality, originality, and usability/security, producing a structured EVAL_REP
Skill Creator
Create, improve, and evaluate Agent Skills with a guided workflow: capture intent, draft SKILL.md, run evals and benchmarks, and optimize triggering description