Development MCP Servers

Turn an API, database or documents into a hosted MCP server that Claude, ChatGPT, Cursor and VS Code can use. No code and nothing to host; an agent can create one from a public API spec with no account.

13 free utility tools for AI agents: web text extraction, markdown/HTML conversion, hashing, encoding, UUIDs, date math, unit conversion, JSON formatting.

Lets coding agents read FlyTrap mobile QA crawl results: bugs with screenshots and logs, screen graphs, network hosts, SDKs, permissions and build diffs for Android and iOS apps.

Scan terms of service for risky clauses before your agent accepts them. Risk scores, pass/flag/block verdicts, quote-grounded clause extraction, change monitoring with webhook alerts, plus a public ToS risk leaderboard. 12 tools, stdio + streamable HTTP.

Lets an AI assistant build levels for Miliastra Wonderland, the UGC editor of Genshin Impact: place scenes from a catalogue of 2,468 official props, write node graphs in TypeScript, then playtest automatically and read back logs and screenshots. Windows only.

Java semantic intelligence for MCP clients and coding agents, powered by Eclipse JDT Language Server and ECJ. Provides code navigation, diagnostics, compilation, JUnit testing, debugging, coverage, affected-test discovery, code actions, and safe refactoring previews without modifying source files.

Search, rank and compare open-source AI projects (self-hosted AI apps, coding agents, MCP servers) with scores, GitHub facts and open-source picks for paid tools.

lets AI assistants explore GitHub repositories, source code, issues, pull requests and commits through 14 validated, read-only tools, plus 4 opt-in write tools that are off by default.

Paid MCP connector that lets dot read and continue existing local Codex chats and retrieve replies on an online Apple silicon Mac. The setup skill is free; the connection service is proprietary.

Check a CVE or a package version against NVD, OSV.dev, CISA KEV and FIRST EPSS in one answer. Remote MCP, no key.

Self-hostable webhook, email, and DNS inboxes that let AI agents capture callbacks, inspect payloads, and forward events to local handlers.

Remote MCP server for Linkrra's OpenAI-compatible inference API for open-weight models. Agents can list models with live per-token prices, estimate the cost of a job, and run chat completions. Listing models and pricing needs no key. Streamable HTTP endpoint: https://pool.linkrra.com/mcp

Assign Linear, Jira or GitHub issues to an AI coding agent team that plans, builds, reviews and tests pull requests.

Audit any URL for SEO and GEO: seven weighted categories scored into a 0-100, a remediation checklist, and four tools over a hosted Apify endpoint.

Bounded local WPE WebKit runtime verifier for agents with native input, DOM checks, fresh PNG evidence, and deterministic cleanup.

Autonomous Machine-to-Machine Cybersecurity Auditor & Pre-Tool Runtime Guard for AI Agents with x402 Micropayments.

MCP gateway for QA/SDET work: Playwright, filesystem, GitHub, Jira, Postgres and 15+ more MCP servers behind one connection, with per-tool allow/deny and read-only-by-default databases.

Open-source, self-hostable MCP gateway that lets AI agents discover and execute SaaS actions with OAuth/API-key credential management, scoped runtime tokens, and action policies.

Captured web interfaces, screenshots and flows as evidence for AI coding agents.

Remote MCP server for a ScaledThought account (an OpenAI-compatible API for open-weight models that your coding agent sets up itself): API keys, budgets per key, usage and prepaid top-ups. Needs an API key: npx -y scaledthought@latest init --region us

Apple developer documentation for coding agents: doc search with developer.apple.com's own results, full API pages with signatures and OS availability, an offline index of 336,629 pages, and 1,700 WWDC sessions (2014–2026) with code. No API key.

Evaluate supplied facts against approved rules, then read or export account-owned decision records; authenticated business pilot, USD 10 per 1,000 completed records, with no external execution.

Screenshot API for AI agents: capture any public page from 227 countries, with cookie banners and ads hidden, in one request. 50 free credits, no card.

Self-hosted MCP for Linux/Termux: shell, files, memory, tasks, browser and an experimental local vault. OAuth-protected remote access; runs on your own machine.

Contact and fraud intelligence for AI agents. Validate emails and phone numbers, look up IPs and domains, and score sign-up fraud risk, from one API key.

ProofRun: bounded HTTP acceptance checks for agent deployments. $0.01 USDC per pack.

Deploy a built website ZIP from your coding agent and get one client review link where clients comment on the page.

🐘 🏠 🍎 🪟 🐧 - Knowledge graph of any PHP codebase built by static analysis (no LLM, no network): calls, routes, message buses, DDD layers and microservice contracts, with confidence levels, plus impact analysis and feature outlines for coding agents.

Zero-dependency, single-file MCP server for AI agents natively running on Termux (Android), Raspberry Pi and Linux: device status, server health and Android/Termux path resolution. No shell or exec surface.

Outside-in domain checks paid per call in USDC on Base: certificate (CT logs + live probe), DNS, email auth (SPF/DMARC/DKIM/MTA-STS), security headers, expiry.