Hosted tools for agents

MCP servers, one URL away.

A hosted MCP server gives your agent live tools and data without a local install. Browse 3,808 MCP servers, copy the endpoint, and use the one-line installer for your client.

MCP servers
3,808
Categories
15
Client guides
14
Distinct hosts
3,667

Showing 456–480 of 570 servers

Clear filters
RylvoDevelopmentBuild, run and manage AI agents in your Rylvo workspace: bots, prompts, guardrails, knowledge bases, channels, tests and flows.rylvo.comView setupSafeguardDevelopmentHosted MCP server (OAuth) for the Safeguard software supply chain security platform: look up projects, SBOMs, findings and vulnerabilities with EPSS and CISA KEV context.mcp.safeguard.shView setupSanctionsKit developerDevelopmentSanctions API docs and request validation, with account-linked synthetic sandbox testing.www.sanctionskit.comView setupScaledThoughtDevelopmentOpenAI-compatible inference with an MCP server for agents: list models and prices, read budgets and usage, top up credits, and manage keys and account settings. The connection uses a key with inference, read, keys and billing scopes, so the server can create and revoke keys and change budgets. Region-pinned hosts; prepaid credits from $10.api.us.scaledthought.comView setupScan.nowDevelopmentRun passive, read-only website security scans and inspect suspicious public links for phishing signals. Checks include TLS, security headers, DNS and email posture, JavaScript libraries, redirects, and lookalike domains. Website scans require ownership or explicit permission.mcp.scan.nowView setupScreen BrowserDevelopmentNarrated demo and tutorial videos of your web app, from a guide or your coding agent.mcp.screenbrowser.comView setupScrewtape.aiDevelopmentPayment-safety checks for agents: decide whether an agent may spend, catch duplicate invoices, track budgets and return signed receipts. $0.005 per paid call or x402; 3 free calls a day without a key.screwtape.aiView setupSemverKitDevelopmentParse and compare semantic versions, check npm-style ranges (^, ~, x, hyphen, ||), find the max satisfying version from up to 200 candidates and compute the next version, node-semver compatible. Free, with no sign-up and no API key; about 60 checks per minute per IP. Nothing you submit is stored. SemverKit is not affiliated with semver.org, npm, GitHub or the OpenJS Foundation.semverkit.mike-tusa.workers.devView setupSentarionDevelopmentGoverned multi-agent orchestration for AI coding agents: a policy gate, a tamper-evident audit chain and the Claude Code Seatbelt. The cited repository returns 404 and the published endpoint returns 502, so the service is unavailable and this listing is retained as a reference.arkhive.dondatabrain.comView setupSEO+GEO AuditDevelopmentAudit any URL for traditional SEO and GEO visibility: a 0-100 score, per-category scores, prioritized findings and a remediation checklist. Four tools (audit_url, audit_deep, get_checklist, report_issue) over a hosted Apify endpoint; $0.50 per 1,000 audit_url calls.mcp.apify.comView setupSequential (usesequential.com)DevelopmentSession replay and product analytics for your agent: the problems users hit (failed requests, JS errors, rage clicks), sessions, funnels, retention.api.usesequential.comView setupSettlensDevelopmentEvaluate x402 V2 payment requirements against explicit policies. Six public offline MCP tools; no wallet, signing or payments.settlens.comView setupShiny GenDevelopmentAI game maker: build, run, screenshot and play-test Godot and retro console games in a live projectmcp.shinygen.aiView setupShipLatchDevelopmentShipLatch runs a free, passive, read-only security scan of a public web app you own, such as a Lovable, Bolt, Replit or Cursor app. Ask your AI assistant to scan a URL and get a grade, issue counts and the type and severity of each issue. No account or API key needed. Streamable HTTP.shiplatch.comView setupShipSafeDevelopmentIndependent security review for AI-built apps: exposed secrets, broken auth, unsafe data access.ship-safe.coView setupShotkilnDevelopmentCapture web pages as PNG, JPEG or WebP, render a URL or HTML to PDF, and generate 1200x630 OG images. REST API and remote MCP server. Free tier.shotkiln.comView setupSiteCheck Agent Intel DevDevelopment135 pay-per-call agent intelligence, token risk, developer and AI data tools for AI agents: prompt-injection and scam filters, PII redaction, token due diligence, OCR and speech-to-text, sports data, app stores and developer data. Paid per call with x402 (USDC on 11 networks, or RLUSD/XRP); no signup, no API key, a failed call is not charged. Part of SiteCheck: 9 MCP servers, 1,194 tools.api.fadeandco.workers.devView setupSiteDoctor - Website Health, SEO Security ScannerDevelopmentInstant website audits: SEO, speed, security headers, SSL, mobile and broken links. Free web scan; pay-per-call MCP tools for agents.manhourplan.comView setupSiteKrsDevelopmentSiteKrs — from 0 to a static website: zero learning curve, save 90% of your time, one-time payment, permanently maintenance-free. The SiteKrs MCP server only lists templates and returns editor links; to use your own source, import a folder/ZIP manually through the “Import Template” dialog on the SiteKrs homepage. / SiteKrs — 从 0 到静态网站,零门槛操作、节省90%时间、一次性付费、永久免维护。SiteKrs MCP 只展示模板并返回编辑器链接,自有源码可在SiteKrs首页「导入模版」用文件夹/ZIP 手动导入。sitekrs.ai-krs.comView setupSitelemetryDevelopmentAuthorized website audits from your agent: security posture, technical SEO, AI visibility (GEO/AEO), analytics integrations, WCAG 2.2 accessibility and performance, returning evidence-backed findings with fixes in 9 languages. Remote Streamable HTTP MCP server with browser OAuth (PKCE + dynamic client registration); no API key to paste. Only audit sites you own or are authorized to test.sitelemetry.comView setupSkillAgentDevelopmentRemote MCP server and directory for AI agent skills, rules files (SKILL.md, CLAUDE.md, AGENTS.md, .cursorrules) and MCP servers indexed from public GitHub repositories. Search, get a skill, recommend skills for a project, and retrieve exact install steps.skillagent.devView setupskilllintDevelopmentLints a SKILL.md before you install it: frontmatter, triggers, secrets, prompt injection, broken references and unbounded autonomy. Local by default, with a remote MCP server for checks.rta.latView setupSkillsPickerDevelopmentFind open-license agent skills (SKILL.md) for your project. Remote, no auth, read-only.skillspicker.comView setupSnapSweepDevelopmentWeb QA for AI agents: real-browser scans with evidence-backed findings. Before the first scan you need a SnapSweep key, a verified host and enough credits; the server at https://mcp.snapsweep.dev/mcp accepts only MCP protocol revision 2026-07-28, and a default 10-page scan needs 340 credits on the balance.mcp.snapsweep.devView setup

Endpoints and setup links are checked against provider documentation where available. Verify access, scopes, and rate limits with the provider before connecting.