MCP 44

Smart, floating AI assistant combining screen analysis, transcription, and automation in real-time. RED is a smart, floating AI hub that deeply integrates into your workflow, combining screen analysis, real-time transcription, and automation capabilities. It helps users think, organize, and act in real-time without needing screenshots or context switching, leveraging Deep Context (MCP) to securely connect to local systems, documents, and databases. It is primarily built for founders, analysts, students, and knowledge workers.

Open-source personal AI agent workstation with private memory and multi-channel access QwenPaw, formerly known as CoPaw, is an open-source personal AI agent workstation developed by Alibaba's AgentScope team. It uses a local-first architecture so users retain ownership of their data, while supporting both local and cloud deployment. QwenPaw includes the ReMe self-evolving memory system, connects to more than 10 communication channels such as DingTalk, Feishu, QQ, Discord, Telegram, and WeChat, and allows one instance to host multiple agents with independent memories. Its Skills and MCP extensions support tool use, agent collaboration, and memory management.

Searchable computer history for people and AI assistants. Screenpipe records screen and audio history locally, making past work searchable. Recall conversations, review how you spent your time across apps, and give AI assistants context through MCP and a local API. Scheduled workflows can turn captured history into meeting notes and work summaries. Screenpipe runs on macOS, Windows, and Linux. The source is available under the Screenpipe Commercial License. Cloud AI, transcription, sync, and integrations can send relevant context off-device when configured.

Largest library of 90,000+ Agent Skills with real-world Use Cases & Community Reviews. Access 90,000+ verified Claude Agent Skills designed for high-performance automation. Every skill includes detailed use cases and peer reviews to ensure reliability. Stop prompting from scratch and deploy expert-level AI modules in seconds.

Agile project management with kanban boards, epics, roadmap, retrospectives, whiteboards and a wiki in one workspace — plus an MCP server, so your AI assistant can run all of it. AgileHero is the all-in-one project tool for agile teams: kanban boards, epics and roadmap, zero-setup retrospectives, estimation poker on the card, an infinite whiteboard, a built-in wiki, and delivery metrics — one product instead of five subscriptions. The part teams notice first: the modules are connected. One click turns whiteboard sticky notes into cards, frames into epics, and a whole mind map into an epic with cards — on your own board, not exported into a second product. Retro action items are already cards. Epics carry a color that follows their work across board, backlog, and search. Everything updates live for the whole team — no asking a colleague to refresh the page. And it's built for the agent era: connect an AI assistant over MCP — it signs in with your account from the client (OAuth), no API key to paste — and let it create, estimate, organize, move and comment on cards and epics. 49 tools cover boards, epics, roadmap, retrospectives, whiteboards, wiki and metrics, bounded by your own permissions, on every plan including Free. Pricing is flat and small-team-friendly: 2 seats free forever, Pro at $12/month plus $9 per extra seat, and a 14-day trial of every premium module — no credit card.

AI-powered platform for website creation and optimization using natural language input. WWWAI.site is a platform that allows you to create a website using natural language input and Claude-powered AI agents. It's a website analysis and optimization tool powered by AI technology. The platform uses specialized AI agents to handle different aspects of website creation, including code creation, requirement analysis, concept setting, and error validation. It also leverages Claude API for advanced language processing and Model Context Protocol (MCP) to ensure consistency across all components.

Shared company brain for AI agents with live context. N71 is a company brain for knowledge workers and AI agents. It connects your tools, maintains a living knowledge graph, and keeps context current across chats and agents through MCP. It helps teams orchestrate multiple agents, surface proactive updates, preserve provenance, and apply governance and permissions so everyone works from the same up-to-date source of truth.

Aident Loadout is the unified tool connecting AI agents to 1,000+ tools with 27k+ actions through one reusable setup. Connect once, use your tools across agents, see every call in Audit, use credits on eligible Aident-managed tools, and add free skills from the Skill Library. Aident Loadout is the unified tool giving AI agents access to more than 1,000 tools with 27k+ actions without making you rebuild the same integrations for every agent. Connect your accounts once through OAuth or Aident Vault, then reuse that setup across Codex, Claude Code, Cursor, ChatGPT, MCP clients, and custom agents through MCP or the Aident CLI. Agents can find the right action for a task, ask for approval when needed, and run it through your approved connection. Loadout Audit shows which tool was called, where the request came from, whether it succeeded, how long it took, and any Aident credit use. Eligible Aident-managed tools can be used with credits, without adding a separate provider key. The free Skill Library gives agents reusable instructions for common workflows, so you can start with the work you want done instead of configuring every tool again.

Desktop AI chatbot supporting multiple LLMs with local data storage. ChatWise is a desktop AI chatbot application that supports multiple Large Language Models (LLMs) including ChatGPT, DeepSeek, Claude, Gemini, and Llama. It emphasizes performance, privacy (storing data locally), and simplicity. ChatWise allows users to chat with various media types like audio, PDF, images, and text files. It also offers web search capabilities and can connect to other tools like Notion and Google Sheets through MCP.

Persistent isolated browsers for scalable AI agent automation Scalebrowser is browser infrastructure for AI agents, providing isolated browser profiles with persistent identities, stored logins, passkeys, cookies, two-factor authentication, captcha handling, and browser automation. Agents can connect through MCP, Python or Node SDKs, REST, or direct CDP while credentials remain encrypted on the user's machine. Each profile preserves its identity across restarts, allowing agents to stay signed in and operate websites repeatedly at scale.

Web-based developer tool to inspect, test, debug, and develop MCP server integrations. MCP Playground is a web-based developer tool designed to inspect and test Model Context Protocol (MCP) servers. It provides an interactive environment for exploring tools, resources, and prompts exposed by MCP servers, making it easy to debug and develop MCP integrations.

Platform for APIs and MCP servers for traditional and AI applications. API Hub is a platform that provides access to enterprise-grade APIs and ready-to-use Model Context Protocol (MCP) servers for both traditional applications and AI agents. It connects users to powerful APIs with cloud infrastructure and predictable pricing. The platform also serves as a central hub for publishing and managing APIs, offering tools for AI integration, documentation, and a marketplace for developers and AI agents to discover and utilize APIs.

Universal MCP for AI chat to act across 600+ apps. Rube is a universal Multi-Client Platform (MCP) developed by Composio that enables users to take actions across 600+ applications directly from within their AI chat interfaces, such as Cursor, Claude, or VS Code. It seamlessly and securely manages authentication and tool selection within the chat environment, allowing AI to interact with a wide range of apps without additional setup.

AI-native platform for orchestrating governed software delivery Skaftor is an AI-native software delivery platform that connects existing development tools, captures product and technical intent, converts requirements into architecture and executable work orders, and orchestrates coding agents through review and deployment. It provides organizational memory, knowledge graph connections, impact analysis, duplicate detection, governance controls, approval workflows, audit trails, and traceability from requirements to shipped code.

AI agent that completes complex, long-running tasks like a human Skygen is a frontier AI agent designed to execute complex, multi-step, and long-running tasks end-to-end. It combines computer-use (vision-based interaction) with powerful backend API tools, enabling it to outperform traditional agents in both speed and token efficiency. Unlike black-box automation, Skygen lets users observe every step and collaborate with the agent in real time

An all-in-one AI coding desktop app for orchestration, QA, and parallel agent delegation. Clean Agent is a comprehensive native desktop application designed as an AI Coding Command Center. it integrates a chat interface, terminal, file tree, in-app browser, and parallel sub-agents into a single workflow. The platform allows developers to automate testing, QA, and coding tasks by plugging in their own AI subscriptions like Claude Code or OpenAI Codex. It features deep research capabilities through an agent-readable browser, context optimization for semantic search, and the ability to delegate tasks to multiple sub-agents simultaneously.

The First MCP-Native Anti-Detect Browser for AI Agents RoxyBrowser is the world's first AI-native anti-detect browser designed to bridge the gap between AI logic and web execution. While LLMs provide the "brain," RoxyBrowser provides the secure "body" needed to interact with the web without being detected as a bot.

Enables Claude AI to query Twitter, Instagram, TikTok, and Reddit data using natural language. Xpoz MCP (Model Context Protocol) is a social media intelligence server that allows LLMs like Claude AI and ChatGPT to query live data from Twitter/X, Instagram, TikTok, and Reddit using natural language queries. It serves as an alternative to traditional, complex, and costly APIs, bypassing rate limits and API key requirements. Xpoz enables AI agents to perform tasks like social listening, sentiment analysis, competitive analysis, and threat detection using structured social data drawn from over 1.5 billion indexed posts.

Persistent context layer for AI agents to connect and reason across business tools. Weavable is a persistent context layer designed for AI agents, connecting business tools like HubSpot, Jira, and Slack into a unified graph. Unlike standard Model Context Protocol (MCP) connections that provide point-in-time snapshots, Weavable maintains a continuous changelog and resolves entities across different platforms. This allows AI agents to reason more accurately with pre-processed, scoped, and ranked data, leading to significantly lower token usage and more reliable, deterministic outputs across real business workflows.

Speclint scores your specs 0-100 across 5 dimensions before any AI agent touches them. Deterministic regex scoring — no AI hallucination in the diagnosis. CLI, GitHub Action, and MCP Server. Speclint is an open-source spec linter that scores user stories and requirements on a 0-100 scale across 5 dimensions: Measurable Outcomes (20pts), Testable Criteria (25pts), Constraints (20pts), No Vague Verbs (20pts), and Verification Steps (15pts). The scoring is fully deterministic — pure regex pattern matching, no LLM involved. AI is only used for the optional rewrite feature, which rewrites failing specs to pass all checks. Available as: CLI (npx speclint lint), GitHub Action, MCP Server (3 tools), and REST API. Free tier: 5 lints/day. Paid plans start at $9/mo for unlimited linting and rewrites. Open source (MIT). Built for teams shipping to AI coding agents — if your spec scores below 70, the agent will hallucinate. Fix the spec first.

Discover product opportunities from payment and tipping signals across 130,000+ websites, then promote what you build with 1,000+ high-DR backlink resources, MCP, and API access. Build or Skip helps indie developers, solopreneurs, and startup teams decide what to build next using commercial signals instead of brainstorming from scratch. Explore 130,000+ websites through revenue rankings estimated from observed payment traffic, dedicated revenue rankings for newly registered domains, and public tipping data. Filter by product category, domain registration date, Domain Rating (DR), and payment channel to uncover smaller or emerging products worth studying. When you're ready to promote, browse 1,000+ high-DR backlink resources, including website submission and comment-posting opportunities. Find direct submission links and check listed fees or reciprocal-link requirements without researching every channel from scratch. Connect your AI assistant through MCP to search and filter products, or use the API to access DR, domain registration details, traffic estimates, and homepage data in your own workflows. Start with evidence, narrow the field, and find a direction worth validating.

Data integration platform with two-way sync, consolidation, and AI-powered automation. Boost.space is a data integration platform that offers two-way sync, consolidation, and standardization of data. It aims to merge diverse data effortlessly, ensuring real-time accuracy and uniformity across various tools. It is the first AI-Ready Data Sync Platform powered by Model Context Protocol (MCP), allowing users to sync, enrich, and automate across 2,000+ integrations without code. It is designed for enterprise-grade, scalable AI success.

Turn meetings into completed client work, in a flash. Supernormal helps agencies and client-facing teams transform meeting recordings into ready-to-send deliverables without a bot on the call. The desktop app captures meetings without a bot joining your call, then the AI agent in the web app generates campaign briefs, pitch decks, proposals, project plans, follow-up emails, and presentations that sound like you. Stop doing busywork and start reviewing polished client work instead.

Knowledge graph for precise AI coding-agent codebase context GitNexus is an open-source knowledge graph engine by Akon Labs that indexes codebases into a deterministic graph of dependencies, callers, imports, call chains, execution flows, clusters, and cross-repository relationships. It connects to coding agents through MCP, allowing them to query structured code context instead of relying only on file searches, embeddings, or similarity guesses. GitNexus supports impact analysis, blast-radius detection, git-diff analysis, architecture discovery, hybrid search, and multi-repository graph analysis. Its local version keeps code, indexing, and queries on the user's machine, while Enterprise provides managed SaaS or self-hosted deployments with features such as automated PR review, auto-reindexing, unified multi-repo graphs, SSO, and dedicated support.