Local-first memory for coding agents that turns coding sessions into reusable engineering memory: decisions, gotchas and domain knowledge.
Agent Orchestrator-first maximum-autonomy worker harness for Codex and Claude Code with evidence-driven verification, typed runtime contracts, and MCP tools.
Adversarial Quality Gate MCP for vibe-coding: IDE AI vs Critic AI, human decides on deadlock
Scan codebases for LLM/AI SDK usage, exposed API tokens, and hardcoded secrets.
Ply MCP server — stdio or Streamable HTTP (--http). List, search, init, doctor, diff, update, and add Angular + Tailwind components (Pro with PLY_LICENSE_KEY). Not MUI Base UI.
Zero-dependency System One decision harness and token optimizer for AI coding agents (OpenCode, Cursor, Claude Code).
MCP server that turns an AI agent's coding session into a reviewable Intent Document — a two-agent author/reviewer distillation, gatekept by a deterministic validator. Works with Claude Code and Codex
Design System Runtime — DSR kernel, DSQL, and lint surface for design token governance
WebSocket-based Model Context Protocol broker. Aggregates multiple MCP providers behind a single endpoint with stdio, SSE, and Streamable HTTP client transports.
Workspace-scoped local MCP execution gateway for AI web interfaces with policy, recovery, and audit controls.
MCP server for the Medusa v2 Admin API – orders, payments, returns, order edits, products, inventory, promotions, bulk changes, sales/customer/restock analytics with interactive MCP Apps dashboards. s
Local-first ADE CLI for planned, verified, reviewable code changes.
Local-first orchestration for coding agents, defined inside your project.
MCP Server for accessing Jamf Documentation (learn.jamf.com)
Self-hosted analytics for MCP servers: which tools, resources and prompts get used, by which client, how fast, and why they fail.
Let Claude Code drive your real, logged-in Chrome. A zero-dependency MCP server + a Chrome MV3 extension that uses element-table perception over CDP. The calling agent is the policy: no second model,
Code review for your coding agent's changes: a cloud bug hunt. From $4; $0 if clean or minor-only.
Open-source broker connectivity for code, apps, and AI agents. 22 brokers, one normalized schema, zero dependencies. Your keys never leave your machine.
Indexes codebase into a local knowledge graph with 40+ AI tools and codebase-guard.
Lockpick from the terminal: device login, project status, findings, repair briefs, Pulse scans and an MCP server for coding agents
MCP server for GSD Task Manager: 20 tools for task CRUD, analytics, and diagnostics over PocketBase. Validated inputs, dry-run support, JWT-aware token health, and dependency-safe deletes.
MCP server for little big brain: search, graph queries, and data import
Shared substrate for Archstone emitters: IR indexing, semantic-type to JSON Schema lowering, and response mapping. No MCP SDK, no filesystem.
MCP server for AI-assisted Mansions of Madness scenario creation with Valkyrie
Deploy private apps and share them by email, from a terminal or an AI agent
MCP server for Manifest provider (Fred) operations (deploy, status, logs, restart, update)
Curated MCP tools from OpenAPI, websites, apps, Postman, cURL or SQL; your agent designs them.
Find and clean leftover processes from AI coding agents on macOS and Linux, including WSL2.
Build a playable game with your coding agent and publish it on Sekai in one step.
Universal Context & Handoff Bridge for AI Coding Tools (IDEs, VSCode extensions, CLIs, and MCP)
Lightweight AI gateway daemon orchestrating Claude Code, Codex, Grok, and Antigravity CLI
GitHub Copilot (VSCode) host adapter for agent-harness-generator. Emits .vscode/mcp.json for the Copilot MCP client (ADR-032).
The wuapi command line: log in from the browser, link a WhatsApp number through a link the person opens (QR or pairing code), run your app with the key, send messages, set up the MCP server and call e
三角洲行动核心数据与随机抽签库(干员 / 武器 / 装备 / 子弹)
Model Context Protocol (stdio) server for Homespun: lets any MCP client (Claude Desktop, Cursor, …) deploy a multi-user web app with hosting, auth, a shared database and permissions included.
On-device, self-organizing memory for AI coding agents — recalls the right facts into each prompt.
完整目錄
IDE、git、程式碼檢查與程式碼分析。共 8,882 個伺服器,最常用的排在前面。每個都會開啟發布者的頁面。