Looking for an alternative to shenmintao/marginalia? Here are the top MCP Server agents Hlido has independently tested — scored on the same framework, ranked by evidence.
shenmintao/marginalia — A library-science-inspired personal knowledge management system with LLM agents. Repo-surface review: measured from the public repository (211 stars, AGPL-3.0, last push 2026-07-08); not hands-on tested.
Read the full shenmintao/marginalia review ›
Hello-QM/catgo-LRG — AI-driven workbench for computational materials science — interactive 3D structure viewer, natural-language CatBot assistant, visual DAG workflow engine, HPC job submission. Tauri desktop app with SvelteKit frontend and FastAPI Python backend. Drives VASP, ORCA, CP2K, QE, GPAW, DFTB+, SIESTA, LAMMPS.. Repo-surface review: measured from the public repository (166 stars, AGPL-3.0, last push 2026-07-15); not hands-on tested.
longyunfeigu/learn-hermes-agent — A 27-chapter hands-on tutorial for building an autonomous AI agent from zero in Python. Agent loop, tool system, memory, skills, MCP, multi-platform gateway, and self-evolution — inspired by Hermes Agent.. Repo-surface review: measured from the public repository (182 stars, MIT, last push 2026-05-14); not hands-on tested.
Oshayr/LLM-Wiki — Autonomous knowledge base plugin for Claude Code - captures reserch, ideas, and decisions into an interlinked wiki with reserch-on-miss, semantic search, and a Wikipedia-style web UI. Knowledge compounds as you work.. Repo-surface review: measured from the public repository (51 stars, MIT, last push 2026-04-26); not hands-on tested.
wanshuiyin/Auto-claude-code-research-in-sleep — ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.. Repo-surface review: measured from the public repository (13480 stars, MIT, last push 2026-07-14); not hands-on tested.
zhitongblog/solomd — A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.. Repo-surface review: measured from the public repository (442 stars, MIT, last push 2026-07-12); not hands-on tested.
skwallace36/Pepper — iOS dynamic library MCP for agents. Repo-surface review: measured from the public repository (102 stars, MIT, last push 2026-04-29); not hands-on tested.
nduckmink/arkon — Arkon: Enterprise AI Knowledge Hub & MCP Server. Self-hosted knowledge base for teams to manage RAG contexts, access policies, and AI skills. Connect Claude and other LLMs via Model Context Protocol (MCP) for automated, secure organizational knowledge integration.. Repo-surface review: measured from the public repository (1062 stars, NOASSERTION, last push 2026-06-03); not hands-on tested.
jordanrendric/claude-video-vision — Give Claude the ability to watch and understand videos — Claude Code plugin with frame extraction and multimodal audio analysis. Repo-surface review: measured from the public repository (997 stars, MIT, last push 2026-07-01); not hands-on tested.
Every score is derived from a fixed 5-dimension framework with C2PA-signed evidence captured during testing. We don't accept payment for placement, so a higher-ranked alternative earned it on the evidence — not on a sponsorship.