Every tool an agent can call

Astro Docs

Web framework for content-driven sites. Search documentation, explore component APIs, and reference framework guides.

AstroFabric

hello-vlrx

AstroFabric is agentic AI for business intelligence. Autonomous AI agents identify the companies, find and verify the people, attach the buying signals, score the fit and deliver finished lists and audiences into the CRM, outreach tools and ad accounts your team already runs. Every workflow carries evidence, approval gates and credit ceilings.

Astryke Hub

selfdefiant007

Every project you have going, on one board your coding agents keep current and you actually read. Other project-memory tools store context for the agent. This one is a board you open. Your agents read it at the start of a session and write back at the end, and when one hits a decision only you can make, it lands on your board instead of stalling the session. - **get_brief** at the start: what you're working on, what each project is waiting on, and which decisions are already settled - **report_progress** at the end: what happened, written by the agent rather than by you - **ask_question** when it hits a fork only you can settle - **answer_thought** to reply to notes you pinned from your phone Works with Claude Code, Codex, VS Code with Copilot, Cursor and Devin Desktop. Auth is OAuth, so there is no API key to paste and no secret on disk. 14 days free, no card. Then $12/month. Setup guides: https://astryke.com/docs

Asyntai – AI Support Agent for Websites

asyntai

Asyntai is an AI support agent that answers your website visitors from your own content. This connector gives you the whole product in chat: add knowledge, edit the agent's instructions, read conversations and leads, and reply to a live visitor yourself, without opening a dashboard.

Atako MCP

romain-laodicina

86 tools grouped in toolsets (?toolsets=), read or read/write API keys

Athenic

maxbeech

Transform Your AI Assistant into a True Business Automation Platform Athenic is the only MCP server that lets ChatGPT, Claude, and other AI assistants actually do the work — autonomously running complex, multi-step business workflows on your behalf. ⸻ Why Athenic? Your Assistant Can Finally Take Action Most MCP servers only handle basic data lookups. Athenic adds an AI workforce to your assistant — so instead of just telling you what to do, it gets it done. What Sets Athenic Apart 🚀 Long-Running Autonomous Tasks Athenic runs for hours, not minutes. Ask your assistant to “research my top 10 competitors and prepare a report” — and return to find a full, sourced analysis waiting for you. 🤖 Multi-Agent Coordination Athenic orchestrates multiple AI agents — research, marketing, analytics — working together to complete full workflows end-to-end. 🎯 Organization-Specific Intelligence It learns your business. Connect your CRM, social accounts, and internal docs so when you ask “create a social campaign,” it already knows your voice, audience, and channels. ⚡ Execution, Not Just Advice Post content, schedule tasks, publish blogs, analyze results — all automatically, with optional approval steps for sensitive actions. ⸻ Use Cases For Startup Founders • “Build our social presence” → Athenic creates, schedules, and optimizes posts across Twitter, LinkedIn, and Bluesky. • “Research our market” → It identifies competitors, finds potential partners, drafts outreach, and tracks responses. For Growth Marketers • “Optimize our SEO” → Athenic researches trends, writes optimized content, builds links, and monitors rankings. • “Run a competitive analysis” → It tracks competitors’ pricing, launches, and campaigns — and alerts you to opportunities. For Operations Teams • “Every Monday, send me a CRM performance report” → Automated data pulls, analysis, and summaries. • “Analyze support tickets for patterns” → Finds recurring issues and suggests improvements automatically. ⸻ Key Features 🔗 Universal Assistant Integration Works seamlessly with: • ChatGPT (Apps SDK) • Claude Desktop (native MCP) • Cursor (development workflows) • Any MCP-compatible client 🛠️ Dynamic Tool Ecosystem Athenic automatically generates tools from your connected services: • Research (PubMed, arXiv, web search) • Social Media (LinkedIn, Twitter, Bluesky) • Business Tools (CRM, analytics, email) • Knowledge Base (RAG-powered queries) 📊 Live Progress Tracking See every workflow unfold in real time — progress bars, execution steps, results, and error logs — all within your AI interface. 🔐 Enterprise-Grade Security API Key, OAuth 2.1, or Anonymous access. Encrypted data, SOC 2–compliant infrastructure, and organization-level isolation. ⸻ Real-World Scenarios Content Marketing: “Write a blog post about AI in healthcare, optimize it, and promote it.” Athenic researches, drafts, schedules posts, and tracks engagement — entirely autonomously. Competitive Intelligence: “Monitor our top competitors and alert me to major changes.” Athenic scrapes, analyzes, and summarizes — only pinging you when it matters. Partnership Outreach: “Find 20 YouTube creators for collaboration.” Athenic discovers, qualifies, contacts, and tracks creators — from start to signed deal. ⸻ For Developers • Protocol: MCP 2024-11-05 • Transport: Streamable HTTP (long-running ops) • Auth: OAuth 2.1, API Keys, Anonymous • UI: ChatGPT Apps SDK + Skybridge • Realtime: Supabase Realtime Example Config: { "mcpServers": { "athenic": { "url": "https://getathenic.com/api/mcp", "transport": "streamable-http", "headers": { "X-API-Key": "ath_live_your_key_here" } } } } One config gives your assistant the ability to orchestrate workflows, access tools, run multi-agent systems, and query your org’s private data. ⸻ Pricing Free Tier: 20 credits, all integrations, community support. Pro: 1,000+ credits, advanced workflows, priority support. Enterprise: Dedicated infrastructure, SLAs, custom agents, white-label options. ⸻ Get Started

ATKN (Agent Trust & Knowledge Network)

jlazoff

Trust-and-knowledge MCP server for agent ecosystems: score trust, verify/certify agents, register identities, and query a Neo4j-backed knowledge graph.

Atlas

centraloperations

Connect AI agents to Atlas MCP for live market data, options analytics, strategy workflows, and 50+ broker-aware trading capabilities.

Atlas Scout

zaguanlabs

Local, read-only structural code navigation for AI coding agents. Build a disposable SQLite symbol index and retrieve definitions, exact source ranges, references, static call paths, architecture summaries, and bounded edit-impact evidence without uploading source code.

AtlasFetch

atlasfetch

Reverse geocode a coordinate to the country, region and municipality that contain it, with ISO 3166 codes, and match it against your own GeoJSON geofences. Free demo key, no signup needed.

Atmita

atmita

Message your Atmita agent - one connection reaches all its accounts and tools, WhatsApp included.

Atmita

support-00bo

Message your Atmita agent - one connection reaches all its accounts and tools, WhatsApp included.

ATOM Pricing Intelligence

a7om

ATOM is the Global Price Benchmark for AI Inference, delivered as a native tool for AI agents. Query live pricing across the inference market through the AIPI (ATOM Inference Price Index) family of benchmarks. Search models, compare cross-vendor pricing, and access market intelligence. Independent and transparent. Free tier included, no API key needed.

Atomic Mail

atomicmail

Email infrastructure for AI agents. Your agent gets a real mailbox and can send, receive and hold a thread on its own. The agent solves a proof-of-work challenge and has a working `@atomicmail.ai` address in about 30 seconds. No signup form, no CAPTCHA, nobody pasting credentials in for it. If the inbox should belong to a person instead, sign in with Google or GitHub over OAuth 2.0, or use a bearer API key for headless setups. **What the agent can do** - Send and receive mail, create drafts, manage threads - Full-text search across the mailbox - Poll the inbox and run scheduled jobs - Pull one-time passcodes and magic links out of incoming mail, so an agent that hits a verification wall gets through it Built on JMAP (RFC 8620 and 8621) rather than a bespoke REST API. Models already know the protocol from training, so they write correct calls without an SDK, and batched method calls go over in one round trip. Three tools cover the whole surface: `register`, `jmap_request`, `help`.

Attestd MCP

attestd

Give your coding assistant or AI agent deterministic CVE and supply-chain signals for any dependency, before it gets added, deployed, or acted on. Returns structured boolean fields (riskState, activelyExploited, supplyChainCompromised) your assistant can state directly or your agent can branch on, no CVSS interpretation, no guessing. Covers 350+ infrastructure products and 256k+ PyPI/npm packages for supply-chain integrity.

Attio CRM

_@kesslerio

Connect AI to your Attio CRM. Manage contacts, companies, deals, and sales pipelines. Create tasks, add notes, and organize lists. Streamline workflows for sales, success, and operations teams.

Audiala

guillaume

Audio guides for 45,000+ places in 1,900+ cities, in 11 languages. Ask about a landmark, museum or city and get a card with a photo, a short story and a 45-second preview of its narrated audio guide. No account or API key needed. **Tools** - `audiala_audio_guide`: one place's guide (summary, photo, 45 s audio preview, link to the full guide) - `audiala_find_city`: a city's must-see places - `audiala_plan_trip`: must-see places for a trip, with a link to Audiala's day-by-day planner - `audiala_contact_team`: contact the Audiala team (asks for your email first) Docs: https://mcp.audiala.com/ · Website: https://audiala.com · Privacy: https://audiala.com/privacypolicy

Audiense Insights

AudienseCo

Extract marketing insights and audience analysis from Audiense reports.

audioknihy-catalog

jan

First Czech audiobook catalog MCP server with cross-partner price comparison. Aggregates Knihy Dobrovský, Martinus, Kosmas, Dobré-knihy, and Radiotéka. Read-only discovery for AI agents — search, browse genres, compare prices, find cheapest offers, lookup author/narrator profiles. 9 tools: search_audiobooks, get_audiobook, find_cheapest_offer, compare_all_offers, browse_genres, search_by_filters, get_author, get_narrator, list_partners. Endpoint: https://audioknihy.cz/api/mcp Discovery: https://audioknihy.cz/.well-known/mcp/server-card.json Docs: https://audioknihy.cz/mcp Transport: JSON-RPC 2.0 over Streamable HTTP, stateless, anonymous. Rate limit: 60 req/min/IP.

Audioscrape

Audio content search and discovery. Find audio tracks, search sound libraries, and browse audio metadata.

AuraCite MCP

m-g-alaeldin

AuraCite measures how AI engines mention, cite, and recommend brands, with an evidence-grade methodology for German and international e-commerce teams. This hosted MCP server exposes that data to any compatible agent: tracked brands, AI mentions, citations, share of voice, visibility score, competitors, trends, per-engine breakdowns, brand comparison, AI-crawler access, and Google Search Console performance (top queries, pages, country and device breakdowns). The server runs at https://auracite.de/mcp/rpc using Streamable HTTP (JSON-RPC 2.0). Authentication is OAuth 2.1 with PKCE S256 and dynamic client registration; an API-key header (Bearer or X-API-Key) works as fallback for CI and headless setups. Default access is read-only (mcp:read). Writes, provider scans, and admin actions are not available in the default scope. They require stronger scopes plus a server-side approval chain with hard credit caps and audit events.

AuraCite Visibility Evidence

auracite

Turn one user-supplied AI answer snapshot into a deterministic incident evidence pack. The result separates the supplied observation, literal calculations, the user's reported concern, limitations, unknowns, a seven-stage assessment, and a same-instrument reproduction protocol. The tool does not browse, fetch or verify sources, call an AI provider, access an AuraCite account, infer causation or stable ranking, or produce a customer-ready verdict. Every reported concern remains unverified and every result keeps external_ready=false until a human completes the evidence review. The upstream AuraCite server requires no AuraCite account, API key, credits, or paid provider call; use through Smithery still requires Smithery's own account or token layer. Do not submit passwords, access tokens, API keys, or unauthorized private customer data. Privacy: https://auracite.de/privacy Support: [email protected]

Auralogs

auralogs

Read-only access to Auralogs production logs and AI analyses. Agents search logs, inspect errors with full stack traces and metadata, and review diagnoses with suggested fixes. 7 tools, no write operations; keys are project-scoped and revocable instantly. Free tier: 10,000 logs/month.

AurasPay Merchant MCP

mazenalshareef33

Connect AI agents to AurasPay merchant workflows with scoped OAuth: review payments, prepare payment links, and require separate human approval for changes.