Data and AI commands

majiayu000/vibeguard

Command Claude Code

Dual-model adversarial review — Claude generates review reports, Codex does adversarial verification, and iterates until convergence.

desek/outlook-local-mcp

Command Claude Code

Run make crud-test from the repository root via the Bash tool. This is the primary, canonical way to execute the MCP CRUD lifecycle test — do not reinterpret the prompt at docs/prompts/mcp-tool-crud-test.md directly. The target wraps scripts/crud-test.sh.

oimiragieo/agent-studio

Command Claude Code

Scaffold a karpathy/autoresearch ML template and launch an autonomous optimization loop.

angelwzr/linux-phone-porting

Command

Iterate research → implement → verify autonomously until a capability works or the model of it is exhausted (linux-phone-porting).

religa/multi_mcp

Command Claude Code

Research the latest available models from Anthropic, OpenAI, and Google, then update the project's model configuration accordingly.

freema/cursor-plugin-cc

Command

Read-only code review of your git diff by a Cursor model. Reports findings; never edits files.

akiojin/unity-mcp-server

Command Claude Code

An interactive release command for projects that develop on a branch named develop and publish changes through a version tag and a pull request to main.

saadshahd/moo.md

Command Claude Code

System instructions for writing effective prompts. Apply when generating commands, skills, agents, or any LLM instructions.

thiswind/cursor-agent-team

Command

You are wearing the /promptengineer mask inside the current Claude Code conversation.

bennypowers/asimonim

Command

Launch a Model Context Protocol (MCP) server for AI-assisted development with design tokens. The server communicates over stdin/stdout using JSON-RPC.

codingthefuturewithai/rag-retriever

Command Claude Code

Interactive CLI command builder and context-sensitive help for RAG Retriever operations.

codingthefuturewithai/rag-retriever

Command Claude Code

Guide users through advanced content ingestion using CLI-only capabilities for rich media, local files, and enterprise integrations.

codingthefuturewithai/rag-retriever

Command Claude Code

Administrative collection operations including deletion, cleanup, and maintenance tasks.

Prorise-cool/prorise-claude-skills

Command Claude Code

Enhanced prompt transformation using session memory and intent analysis with --enhance flag detection.

tumf/conflux

Command

Run Conflux acceptance review (prompt provided by orchestrator).

jkitchin/discopt

Command

Debug a broken discopt setup, model, or solve. Root-cause installation/import failures, daemon problems, infeasibility (with IIS), NLP/numerical errors, wrong answers, non-reproducible runs, and crashes — then fix them. Use when something is broken or behaving unexpectedly, as opposed to /diagnose which interprets a…

jkitchin/discopt

Command

Translate a natural-language optimization problem into a complete, runnable discopt model — variables, objective, constraints, validation, and a solve call. Use when the user describes a problem in words and wants discopt model code.

John-Wendell/Attention-MoA

Command

Command "data_cleaning" from John-Wendell/Attention-MoA, covering data cleaning, requirements, steps, convert html to markdown and keep or remove specific languages.

ml-jku/conglude

Command Claude Code

Score unlabeled ligands against proteins using ConGLUDe. Invoke ONLY when the user explicitly mentions "ConGLUDe" or "conglude" together with prospective prediction intent — e.g. "run predict", "score ligands", "virtual screening", "predict affinity", "run predict.py", "screen a library", "embed a library", "encode…

ml-jku/conglude

Command Claude Code

Train the ConGLUDe model. Invoke ONLY when the user explicitly mentions "ConGLUDe" or "conglude" together with training intent — e.g. "train", "start training", "run train.py", "fine-tune", "retrain", "train the model", "debug training run". Do NOT trigger on evaluation, prediction, or embedding requests — use the…

TheSmokeDev/taskchad-os

Command

Build the dual-variant prompt pack, imagegen packet, and initial manifest using the selected library template.

dtannen/icon-agents

Command Claude Code

Intelligent AI/ML analysis that automatically selects the most relevant legendary data science and AI experts for your project.

conradkoh/raggrep

Command Cursor

Run manual sanity checks on the RAGgrep search system to ensure ranking quality.

redhat-ai-americas/memory-hub

Command Claude Code

Wire MemoryHub integration into the agent for persistent memory across runs. This command configures both the SDK path (self.memory for agent code) and the MCP path (MemoryHub tools for the LLM).