- retriEVAL @hcarrillo001/retrieval-mcpLLM evals as MCP tools: score outputs for faithfulness, relevancy, and hallucination.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.
- GGUF @arose26/gguf-mcpInspect local GGUF/safetensors models: quantization, params, VRAM fit — headers only.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- VMware Private AI (GPU & Model Serving) @vmware-skills/vmware-privateaiVMware Private AI (NVIDIA): GPU inventory, vGPU profiles, utilization, model serving — 17 tools5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Krauncher analyzer @ilya-a-sergeyev-ger/krauncher-mcpHow long a GPU job will run and cost, before running it — static code analysis, never executed.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- GenieLocker @nhantour/genie-lockerDiscover and quote quality-gated private AI inference without creating or buying a locker.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.
- Vibedonate @pooriaarab/vibedonateBitTorrent-style mesh for AI inference across agentic CLIs. Local-first.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- vLLM Ops @jaimenbell/vllm-ops-mcpRead-only ops/health MCP server for a local vLLM instance: health tiers, GPU/VRAM, service status.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Mendrift @suneel190700/mendrift-mcpMCP server for ML drift detection, MLflow registry diffing, and HMAC-gated model rollback5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- TrainTools @aparajeets/traintoolsRecommend paper-backed diagnostics for PyTorch and Hugging Face training problems.2% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- AI Guardian @aiops-tools/ai-guardianGoverned local-LLM observability: model policy, prompt scanner, capture proxy, 21 tools.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Inference AIops @aiops-tools/inference-aiopsGoverned GPU inference ops (vLLM + Ray Serve): latency RCA, scaling, drain, 39 tools.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- LLMTest @tjacquesson/llmtest-mcpBenchmark AI models on real prompts. Find cheaper, faster alternatives across 340+ models.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Telegraph @telegraphprotocol/telegraphDecentralized AI inference (weather, deepfake detection, LLMs) with x402 USDC micropayments.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Qwen 2 @runapi-builder/qwen-2-mcpRunAPI MCP server for Qwen 2: create tasks, poll status, check pricing.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- ML Lab @rohithraju-ops/mcp-ml-labRun end-to-end ML experiments from natural language (XGBoost, LightGBM, Optuna).4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Qwen MCP @rocnubie/chat-mcpQwen AI Chat guide to Qwen3.6-Plus and qwen3.6-flash with practical notes on 1M context, coding7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.
- DeepSeek FR MCP @rocnubie/deepseekfr-mcpIndependent French workspace for evaluating DeepSeek-V3-class chat and DeepSeek-R1 reasoning.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.
- Safe Kaggle MCP @parkseokjune/kaggle-mcpSafety-first Kaggle MCP — confirm-gated destructive ops, submission budget, injection-fenced.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- PathCourse Health (PCH) @pathcourse-health/mcp-serverCheaper LLM inference, USDC agent payments via x402, agent identity / reputation lookup.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Golden Dataset @nipunkhanderia/golden-dataset-mcpVersion-controlled golden datasets and RAG evaluation, no API key needed.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- MyAi @myaitoken/myai-mcpDecentralized AI inference on Base: OpenAI-compatible chat via community GPUs, paid in MYAI9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- redline @gowtham0992/redlineRun local prompt regression checks from prompt-response logs in AI coding assistants.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Offline @gabrielmahia/offline-mcpMCP server enabling offline-first agent operation over intermittent connectivity4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- DCP @dhnpmp-tech/dcp-mcpSaudi sovereign AI compute: OpenAI-compatible inference, rent GPU pods, manage storage.9% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.
