MCP 44

AI Product Management for Product Teams, Agencies, and Startups metastory AI empowers product teams, agencies, and startups to turn ideas into production-ready deliverables. With AI-generated PRDs, user stories, and acceptance criteria, plus seamless UI-to-feature mapping, Figma and Jira integrations, and MCP-powered Meta Coding, it removes ambiguity and accelerates the entire software development lifecycle. Whether collaborating with developers, building on no-code platforms, or guiding AI coding assistants like Claude, Cursor, or Copilot, metastory keeps requirements clear, contextual, and perfectly aligned.

A directory for discovering and sharing Model Context Protocol (MCP) Servers. MCP Servers is a platform that collects and organizes Model Context Protocol (MCP) Servers. It serves as a central directory where users can discover, share, and learn about various MCP Servers available for AI applications. MCP Servers enhance AI capabilities by providing access to external data, tools, and prompts through a client-server architecture, enabling AI systems like Claude to securely connect with various data sources.

Open-source integration infrastructure for AI agents and third-party applications Corsair is an open-source integration layer for AI agents and applications. It provides secure, multi-tenant OAuth authentication, permissions, MCP support, webhooks, credential management, caching, and provider-specific error handling for services such as Gmail, Slack, Notion, HubSpot, Airtable, Google Calendar, Stripe, Linear, GitHub, and Discord. Developers can self-host the SDK for free or use Corsair's managed cloud platform to connect third-party applications without maintaining authentication flows, token rotation, webhook verification, API changes, rate-limit handling, and tenant credential isolation.

Sendmux is an email inbox API for AI agents. It gives agents dedicated mailboxes, multi-provider sending, live inbox events, and developer tools so they can send, receive, route, and react to email in production. Sendmux helps developers give AI agents a real email identity. Instead of stitching together a sending API, mailbox provider, parser, webhook layer, and billing system, teams can use one platform for inbound mailboxes, outbound sending, routing, quotas, webhooks, and live inbox events. Agents can receive email in dedicated mailboxes, read messages and threads, process attachments, send replies, and stream inbox activity through Server-Sent Events. Outbound sending supports SMTP, Gmail OAuth, Outlook OAuth, and Sendmux-managed Amazon SES, with delivery groups, weighted routing, per-provider quotas, health checks, and failover. Sendmux is built for AI agent teams, automation agencies, SaaS platforms, and developers building support agents, sales agents, scheduling agents, monitoring agents, and workflow agents. It also includes REST APIs, OpenAPI specs, SDKs, CLI, MCP tools, and Sendmux Skills.

Omni LLM chat platform for team productivity, integrating multiple AI models in one place. Helpedby.ai is an interface that allows you to interact with multiple LLMs (Large Language Models) in one place. It provides a seamless experience for users to access and utilize various AI models for different tasks. It is an Omni LLM chat platform for team productivity, allowing users to personalize their chat experience with multiple AI models, manage & collaborate with their team, and create their own agents without a dev team. It offers features like AI Agents, Model Router, Pay as You Go pricing, collaboration tools, and the ability to build MCPs and use Agentic RAG.

Dedalus Machines are the fastest persistent sandboxes for AI agents. Spin up long-running, stateful agents that never sleep in <50ms. Dedalus Labs provides cloud infrastructure purpose-built for AI agents that need to run for hours, days, or continuously in production. Unlike ephemeral sandboxes that destroy state after each session, Dedalus Machines give every agent its own persistent Linux microVM. Files, installed packages, environment variables, caches, and running processes all survive across sleep and wake cycles. Sleep is free — you only pay for compute when the agent is actually running. Key capabilities: - 50ms cold start with scale-to-zero (no idle charges) - VM-level hardware isolation for running untrusted LLM-generated code - Persistent POSIX-compliant filesystem across sessions - Per-second billing starting at $0.0504/vCPU-hr - GPU-capable VMs for local model inference - Built-in MCP (Model Context Protocol) support with a hosted marketplace of production-ready MCP servers - Multi-model SDK supporting Claude, GPT, Gemini, DeepSeek, Mistral, and 100+ models - Dedalus Auth (DAuth): zero-trust, multi-tenant authentication for MCP servers Dedalus is designed for developers building autonomous coding agents, research agents, CRM agents, monitoring agents, and any stateful AI workflow. Free tier includes $20 sign-up credit and 50 compute hours per month with no credit card required. Backed by Y Combinator (S25) with $11M in seed funding from Kindred Ventures and Saga Ventures.

Human and AI shoppers experience your website and tell you why they'd leave. AI persona shoppers browse your real site like customers and narrate the friction in their own words; AI shoppers reveal how ChatGPT, Claude & Gemini see and recommend you. Experience report + AI Readiness Score in ~10 minutes. Free tier, no card. Your website has two kinds of customers now — people and AI assistants — and mocktomer shows you how it feels to be each of them. AI persona shoppers (budget hunters, mobile skimmers, impatient buyers) browse your live site the way real customers do: they hesitate where people hesitate, get lost where people get lost, and then tell you, in their own words, why they'd leave. You get a ranked experience report — every friction moment with severity, real shopper quotes, and a plain-English fix — in about 10 minutes. AI shoppers then reveal how ChatGPT, Claude, and Gemini perceive your site: an AI Readiness Score (0–100), what confuses the assistants, and which competitors they recommend instead of you. Mark findings fixed and send the same shoppers back to confirm the experience actually improved. Night Shift keeps shoppers walking your money paths on a schedule, with Slack, Jira, and webhook alarms the moment something that used to work starts failing. A built-in MCP server means Claude, ChatGPT, and Cursor can send shoppers in and read the reports for you. No code, no snippet — paste a URL and meet your customers.

TrustGate by NeuralTrust is the only open source AI gateway with built in security for enterprises running LLMs and AI agents at scale. It sits in front of AI traffic to control access, enforce policies, and block threats like prompt injection and jailbreak attempts before they reach production. TrustGate ships with a catalogue of over 200 MCP connectors, so teams can plug in agents and tools without building custom integrations for every provider. TrustGate is an open source AI gateway built for enterprises deploying LLMs and AI agents in production. It gives security and engineering teams a single control point for AI traffic, with built in protection against prompt injection, jailbreaks, and other agent specific threats. TrustGate includes a catalogue of over 200 MCP connectors, letting teams integrate agents and tools quickly without custom work for each provider. Unlike closed, proprietary gateways, TrustGate gives teams full visibility into how their AI infrastructure works and the flexibility to adapt it to their own security and compliance requirements. It is built for enterprise scale, with the observability, governance, and reliability needed to run AI workloads in regulated industries like banking, aviation, and energy.

AI-powered code editor for developers and enterprises, enhancing productivity and workflow. Windsurf (formerly Codeium) is an AI-powered code editor designed for developers and enterprises. It features the Windsurf Editor, an AI-native IDE that aims to keep developers in a state of flow by anticipating their needs, fixing issues proactively, and handling complex codebases. It offers features like Cascade, an agent that codes, fixes, and thinks ahead, and Windsurf Tab, which provides smarter code suggestions based on command history and clipboard content. Windsurf also supports building, iterating, and shipping apps within the editor, along with features like memory, lint fixing, MCP support, and terminal command integration.

Connect any MCP-compatible AI assistant to Debitura's cross-border debt collection platform. Check case status, read partner conversations, get pricing, and submit new collection cases across 183 countries, directly from Claude, Cursor, or VS Code. Debitura is an international debt collection platform that recovers unpaid B2B and B2C invoices in 183 countries through licensed local collection partners, on a no-cure-no-pay basis. The Debitura MCP server exposes that platform to AI assistants through the Model Context Protocol. Point any MCP-compatible client (Claude web and desktop, Claude Code, Cursor, or VS Code) at the remote server and authenticate with your Debitura API key. Your assistant can then check the status and timeline of collection cases, read messages from local collection partners, retrieve pricing and eligibility estimates before submitting, list and upload case documents, message partners, and open new collection cases. Case creation is safety-wrapped: it previews first, requires explicit human confirmation, and is idempotent, so no duplicate cases are created. It is a remote, streamable-HTTP server with no local install, authenticated with an API key passed in an XApiKey header. Finance and operations teams can manage international debt recovery in natural language without leaving their AI assistant.

Interactive platform for exploring, testing, and integrating MCP APIs. MCP Showcase is an interactive platform that provides a live, no-risk MCP playground for companies to explore, chat with, and integrate their Model Context Protocol (MCP) APIs in minutes. It aims to accelerate evaluation and drive higher integration rates for MCP servers by delighting developers and convincing decision-makers. The platform offers a sandbox environment for testing MCP functionality, interactive documentation, chat features, and real-time usage insights to refine product features and improve API design.

Mobile remote control for local Claude Code AI agents. PUNK is a remote control application designed specifically for Claude Code running on a laptop. It allows users to interact with their local AI agents from their mobile phones, effectively turning the laptop into a 'mainframe' and the phone into a portable 'terminal.' PUNK enables developers to monitor sessions, approve or deny tool calls, and provide instructions via text or voice while away from their desks. The system uses a secure relay model, ensuring that prompts and responses are not stored in a persistent database and that the laptop remains secure by only using outbound connections.

Ecommerce data and AI video tools for product research, market intelligence, and content creation. Nexscope brings multi-platform ecommerce data, market research, and AI video creation into one workspace. Sellers can analyze products, pricing, keywords, demand, competitors, customers, and advertising signals across leading marketplaces, then use those insights to improve product, listing, advertising, and creative decisions. Teams can also turn product assets into product and UGC-style videos. Ecommerce data can be accessed through APIs, connected to compatible AI tools through MCP, or used through focused skills in supported workflows.

Describe a video and Treza builds the AI pipeline that makes it, then runs it on demand, on a schedule, or behind an API. Treza turns a description into a working AI video pipeline. Tell it what you want in plain language, like a daily nature Short for YouTube, vertical clips cut from a podcast, or a product spot from a written brief, and Treza drafts the whole workflow: the script writer, the video and image models, voiceover, music, captions, and the publish step. Run it once, put it on a schedule, or call it from your own code. Every step is a node on a visual canvas, so nothing is a black box. Open any pipeline to swap a model, rewrite a prompt, or add a step, then rerun just that part. Treza includes leading video models such as Veo 3.1, Seedance 2.5, Kling, and Wan, Gemini image models, and open text models like Llama, DeepSeek, and Qwen. Nodes also cover motion transfer, lip sync, upscaling to 4K, background removal, transcription, and text to speech. A built-in timeline editor handles the final cut, with layered tracks, transitions, and an AI assistant that makes edits on request. What you can do with Treza: - Turn a one-line idea into a captioned, scored Short in chat - Clip long videos and podcast episodes into vertical clips, cut cleanly between words - Run a faceless channel on autopilot with scheduled pipelines - Post finished videos straight to connected YouTube and TikTok channels - Call any pipeline as an API, including an OpenAI-compatible endpoint - Build and run videos from Claude over MCP, or with the Treza plugin in ChatGPT Pricing is simple: prepaid credits from $5, no subscription, and credits never expire. You only pay for successful runs, and every run shows what each step cost. For studios, agencies, and brands, Treza Enterprise adds deployment in your own cloud, custom nodes and workflows, and a forward deployed engineer. Built by people from film and visual effects, Treza brings the pipeline thinking of professional post-production to generative video.

Build production-ready apps and websites with enterprise-grade security using Claude or ChatGPT. Apper manages the backend, with no extra Apper AI credits. Apper lets individuals and businesses build production-ready websites and full-stack web apps with enterprise-grade security built in. Use the Claude or ChatGPT subscription you already pay for, choose your preferred model, and build without buying extra Apper AI credits. Describe what you want to create in plain English. Apper connects to your AI through MCP and handles authentication, databases, permissions, billing, integrations and backups. It also handles app hosting and publishing, with an Apper URL or your own custom domain. You can focus on your website or app without assembling and maintaining the backend yourself. Whether you’re a solo founder launching an MVP, a freelancer building for clients, or a business creating internal tools and customer portals, Apper provides the infrastructure to take your idea live. Build business websites, SaaS products, CRM systems, booking apps and more. Start free and upgrade as your data, storage and business needs grow.

One key and one prepaid balance for 2,500+ data APIs from 90+ providers that your AI agent can call over MCP or CLI. Pay per call, price shown before every run. looot gives your AI agent one key and one prepaid balance for 2,500+ data APIs from 90+ providers: Google search results, SEO and backlink data, company and people enrichment, work email finding and verification, phone lookup, web scraping, news and maps. It works as a remote MCP server in Claude, Claude Code, Cursor, Codex, ChatGPT and VS Code, and as a CLI in the terminal. The agent searches the catalog in plain words, sees each endpoint's inputs and price before it runs, and pays per call from one balance. Failed calls are not billed. looot is built by AY Automate.