From 81e3f2b5eb3623d5383de8dd217ea461f23223be Mon Sep 17 00:00:00 2001 From: cm Date: Mon, 11 May 2026 07:29:06 +0900 Subject: [PATCH] Add playidea-lab/pcq to Data Science Tools MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Agent-operable ML experiment contract — cq.yaml + JSON contracts + built-in MCP server exposing 14 tools (resolve_project, inspect_project, run_experiment, validate_run, describe_run, compare_runs, lineage_chain, ...) for running, validating, and tracing experiments across any framework. Apache-2.0. Co-Authored-By: Claude Opus 4.7 (1M context) --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index a58a33b32..7b827526b 100644 --- a/README.md +++ b/README.md @@ -1132,6 +1132,7 @@ Integrations and tools designed to simplify data exploration, analysis and enhan - [mckinsey/vizro-mcp](https://github.com/mckinsey/vizro/tree/main/vizro-mcp) 🎖️ 🐍 🏠 - Tools and templates to create validated and maintainable data charts and dashboards. - [optuna/optuna-mcp](https://github.com/optuna/optuna-mcp) 🎖️ 🐍 🏠 🐧 🍎 - Official MCP server enabling seamless orchestration of hyperparameter search and other optimization tasks with [Optuna](https://optuna.org/). - [Whatsonyourmind/oraclaw](https://github.com/Whatsonyourmind/oraclaw) [![Whatsonyourmind/oraclaw MCP server](https://glama.ai/mcp/servers/Whatsonyourmind/oraclaw/badges/score.svg)](https://glama.ai/mcp/servers/Whatsonyourmind/oraclaw) 📇 ☁️ 🏠 🍎 🪟 🐧 - Decision intelligence MCP server with 19 algorithms (bandits, Monte Carlo, constraint optimization, forecasting, anomaly detection, risk analysis, graph algorithms), 12 MCP tools. Install via `npx -y @oraclaw/mcp-server`. +- [playidea-lab/pcq](https://github.com/playidea-lab/pcq) 🐍 🏠 🍎 🪟 🐧 - Agent-operable ML experiment contract (cq.yaml + JSON contracts) with a built-in MCP server exposing 14 tools (resolve/inspect/run/validate/describe/compare/lineage) for running, validating, and tracing experiments across any framework (PyTorch / HF Trainer / Lightning / sklearn / XGBoost). Apache-2.0. - [phisanti/MCPR](https://github.com/phisanti/MCPR) 🏠 🍎 🪟 🐧 - Model Context Protocol for R: enables AI agents to participate in interactive live R sessions. - [phuongrealmax/code-guardian](https://github.com/phuongrealmax/code-guardian) 📇 🏠 - AI-powered code refactor engine with 80+ MCP tools for code analysis, hotspot detection, complexity metrics, persistent memory, and automated refactoring plans. - [pramod/kaggle](https://github.com/KrishnaPramodParupudi/kaggle-mcp-server) 🐍 - This Kaggle MCP Server makes Kaggle more accessible by letting you browse competitions, leaderboards, models, datasets, and kernels directly within MCP, streamlining discovery for data scientists and developers.