"Less" matching MCP servers:

Enables AI assistants and other MCP clients to launch and steer Cursor Cloud Agents on repositories or as repo-less research tasks, including model selection, mid-run follow-ups, cancellation, result retrieval, and run listing. Also exposes authentication checks and per-agent or per-run token usage and cost reporting.

Provides tools for optimizing, fine-tuning, and deploying large language models with Unsloth, enabling 2x faster training with 80% less memory through model loading, fine-tuning, text generation, and model export capabilities.

MCP server that enforces single-use human confirmations for sensitive tool calls, sanitizes prompts against prompt injection, and re-validates file paths before destructive actions.

An MCP security trust-check server that evaluates component installation risks and provides safe-install decisions for autonomous agents.