Typed host-side helpers for Convex backends — builders, errors, validators, relationships, pagination, HTTP, env, and observability.
TypeScript SDK、核心的預編譯二進位綁定與命令列工具,三者為同一次發布產出
Check whether a backtest is real: deflated Sharpe, CSCV probability of backtest overfitting, minimum track record and backtest length, haircut Sharpe and luck-equivalent trials, as an MCP server. Loca
Production readiness for vibe-coded apps — know your AI code is ready to ship
Comprehensive security guards for LLM-powered and agentic AI applications - 35 guards covering OWASP Top 10 for LLMs 2025, Agentic Applications 2026, and MCP Security. All guards accessible via unifie
Domain security reconnaissance MCP server — 34 tools for AI agents, also usable from the command line (npx dechonet-mcp ssl example.com). Phishing/smishing link checks, external exposure maps, DNS, SS
Production-grade TypeScript SDK for Ollama with native fetch, high availability failover, multi-turn tool calling, structured outputs, reasoning stream tokens, OpenAI/Anthropic bridges, MCP, Web Strea
AI coding config sync CLI. Keep rules, commands, agents, skills, MCP servers, hooks and permissions in one .agentsmesh source and generate native config for Claude Code, Cursor, GitHub Copilot, Gemini
The amdahl command line: sign in, check setup, optimize messages, manage API keys, connect MCP clients.
Comprehensive security sanitization library for Model Context Protocol (MCP) servers with trusted security libraries
Official MCP server for the InternetData API. Lets an LLM agent look up any IP address, AS or domain, see what its key's plan includes, browse every IP database we publish and where an organization's
Record narrated product demos with your coding agent: the tday MCP server and CLI for Claude Code, Cursor, Codex, VS Code and Gemini CLI. Only rendered UI and metadata leave your machine.
Preview real components and native Django, Jinja2, Handlebars, PHP and Ruby templates. Inspect UI edits from coding agents and VS Code without starting the app.
Model Context Protocol (MCP) server for Mealie — exposes every Mealie REST API endpoint as an MCP tool.
MCP tool handler logic for the GGUI protocol. Pure over @ggui-ai/mcp-server-core seams and consumed by @ggui-ai/mcp-server. Never imports AWS, HTTP transports, or CLI concerns.
The `ggui` CLI — run the GGUI protocol locally: dev server, self-hosted MCP server, and gadget / blueprint authoring. End-to-end OSS and self-hosted workflows; no account required.
Making Apple docs AI-readable
Pure-TypeScript registry operations and storage interfaces for the GGUI marketplace. No I/O, no HTTP — operations take typed inputs plus a storage deps bag and return discriminated-union results.
Read any company's record on the Trooth Network from your terminal: each fact with who said it and when, and Trooth's signed witness statement checked on your machine. No key, no account. Claim your c
Prompt injection scanner for documents, web pages and RAG content: instruction overrides, spoofed chat roles, hidden text and data-exfiltration attempts, as a CLI and an MCP server
Experimental: MCP gateway that labels every tool result and blocks risky tool calls using ADTL trust labels
Model Context Protocol server for the FerrLabs platform (orgs, projects, subscriptions, docs, tokens, public stats)
MCP server for semantic search and write actions across Apple Mail, Messages, Calendar, and Contacts. Speaks stdio (local) or bearer-token-authenticated Streamable HTTP (LAN / Tailscale).
Local AI gateway for OpenCode with tool/function calling — use any model via OpenAI, Anthropic, or Gemini API format
Lethal trifecta detector for MCP tool catalogs: flags servers that combine untrusted content, sensitive data and exfiltration or destructive tools
MCP server for OpenStreetMap: geocoding, walking/driving/cycling distances, route optimization, isochrones and POI search
Pin what your MCP servers are allowed to do. Fail CI when it changes.
SurfCAD agent bridge MCP server (stdio) — pairs to a browser tab via the local WebSocket relay, or a saved remote login
Too many MCP servers eating your context? Tool Guardian as a native DeepSeek Harness (DSH) bundle sits in front of all of them and loads each tool's schema only when it's needed. A bridge to the Pytho
Full Court Defense CLI — security scanning for AI agents from your terminal
Agentic Mermaid: render, verify, and safely edit Mermaid diagrams as ASCII, PNG, SVG, or structured agent workflows — with composable styles (hand-drawn, excalidraw, watercolor, or your own as JSON).
AI coding agent for the terminal: staged file edits with diff review, permissioned shell, web search, autonomous mode, sessions, and an MCP client. Zero runtime dependencies.
MCP (Model Context Protocol) adapter extension for Pi coding agent
DSH plugin: import the connections DBeaver already knows (saved passwords included), query them through nine MCP tools across SQLite, PostgreSQL, MySQL, Oracle and openGauss - shared by every DSH prof
MCP server exposing the anonymizer to coding agents: files read through it reach the model masked. It does not stop an agent reading files directly - with a coding agent, use veilio-guard (Claude Code
Agent Plugins 1.0.0 client for the Pi coding agent. Discovers, validates, and loads portable plugins (skills + MCP servers) from plugin.json packages.
Tüm dizin
IDE'ler, git, linting ve kod analizi. 8,501 sunucu, en çok kullanılanlar önce. Her biri yayıncının sayfasını açar.