Best Edge AI Skills & MCP Servers

Your AI agent has read your codebase but still can't safely change it. unerr is a local guardrail that hands the agent the call graph and your rules at the moment it edits.

Information-geometric agent memory with mathematical guarantees. 4-channel retrieval, Fisher-Rao similarity, zero-LLM mode, EU AI Act compliant. Works with Claude, Cursor, Windsurf, and 17+ AI tools.

The local-first falsifiable decision ledger — MCP server that turns an Obsidian vault into a compounding decision system with multi-model AI council dissent and outcome-driven refutation propagation.

AI Memory & Knowledge Management MCP Server — 32 tools, BM25 R@5=95%, hybrid search (RRF), knowledge graph, analytics, cloud sync, Obsidian compatible. Works with Claude Code, Cursor, Gemini CLI, Windsurf, hermes-agent.

Shared, persistent memory for AI assistants, built on the Zettelkasten method.

An LLM Wiki for Claude Code, in your Obsidian vault.

Persistent memory for AI coding agents. Knowledge graph with 4-tier taxonomy, confidence scoring, and MCP protocol support for Claude Code, Cursor, Gemini CLI, Windsurf, Cline, Codex, and OpenCode.

MCP server for ProjectionLab — export, snapshot, and restore financial plans with a built-in FIRE advisor knowledge base.

MCP server exposing knowledge search tools for AI coding assistants

MCP server providing persistent cross-session memory for AI agents — store, recall, search, and summarize memories across agent boundaries

Scaffold a personal LLM wiki (Karpathy pattern). MCP-ready, multi-agent, publishable. Zero-prompt install for IM agents (OpenClaw, Hermes, Codex, Claude Code).

FreeRide — Project knowledge for AI agent workflows. MCP server that gives AI coding agents persistent memory across sessions.

DEPRECATED — install the Claude Code plugin at gramatr/claude-plugin instead. This npm package no longer ships a CLI; Codex and Gemini integrations are paused pending HTTPS+OAuth MCP support.

MCP server for Agent Zone — infrastructure knowledge for AI agents

Node.js/TypeScript MCP server for Atlassian Confluence. Provides tools enabling AI systems (LLMs) to list/get spaces & pages (content formatted as Markdown) and search via CQL. Connects AI seamlessly to Confluence knowledge bases using the standard MCP in

Knowledge base — loads knowledge, fts, graphbuild, graphquery for document indexing and graph-powered search.

Local memory layer for AI agents — coding assistants over MCP, game NPCs over REST

MCP server for Yuque (语雀) — expose Yuque knowledge base to AI assistants via Model Context Protocol

MCP server for Obsidian vaults — read, write, search, and surgically edit notes, tags, and frontmatter via the Local REST API plugin. STDIO or Streamable HTTP.

AGLedger™ SDK — Accountability and audit infrastructure for agentic systems.

MIND MCP Server — 28 tools. The most complete AI memory layer with the Focus → Project → Outcome → Task hierarchy (mind_focuses + extended mind_life with item_type/parent_id/focus_id and cross-account project sharing via share/list_shares/revoke_share), d

GeneXus 18 MCP server — read, edit, and analyze GeneXus knowledge base objects (transactions, web panels, procedures, SDTs) directly from Claude, Cursor, and other AI agents over the Model Context Protocol.

The control plane for your team's AI tools — share MCP servers, knowledge bases, and usage analytics across your engineering team

MCP tools that search, write, and auto-link your Obsidian vault — and learn from your edits.