Development MCP Servers

One OAuth connector URL that gives Claude, ChatGPT and Claude Code your team's remote MCP servers, with Tool Packs, guardrails and a request log.

Finds what an AI job needs, shortlists from 450+ OpenRouter models, and tests the best on your own tasks. Free to start.

Read YouTube video details, descriptions, statistics and available chapters for structured video research. FlashData API calls are paid.

Inspect YouTube Creative Commons license labels for rights review; labels do not grant training permission. FlashData API calls are paid.

Read a YouTube channel profile and recent videos by handle to build a bounded upload inventory. FlashData API calls are paid.

Find the cheapest API relay or provider for any model from a price directory refreshed weekly, plus official and relay free credits. Remote, no sign-up or API key.

Paid starter kit ($39) for building MCP servers and Cursor plugins: TypeScript MCP server (stdio + Streamable HTTP) with client tests and CI, optional Python FastMCP variant, plugin manifest and validator.

Free web tool that generates a ready-to-run MCP server (TypeScript or Python) from a simple form — no OpenAPI spec needed. Download a ZIP and connect it to Claude Desktop, Cursor or any stdio MCP host.

Read-only Playwright CI results for coding agents: tells whether a failing test is new in this PR, already failing on main, or known flaky, with per-test history.

Remote no-auth MCP that recommends and compares vibe coding tools and AI coding agents: fit, pricing, free tiers, alternatives, demo videos.

A self-hosted hub for your Android phones. Let Claude Code, Codex or any MCP client use them with approvals and an audit log, or take control yourself.

Open automation engine for AI agents. Control live browsers and desktop apps locally.

Outside checks for the sites your agent ships: an A to F health grade for HTTP headers, TLS, DNS and cookies, with fix steps. Tools: check_site, list_sites, get_site.

Let your AI run builds, tests and dev servers in the folders you pick. Background processes, Git, allowed commands and localhost checks. Stops everyday accidents; not a sandbox.

Starter kit for small businesses: A2A agent card, MCP server card, minimal example server and validator. Python standard library only, MIT, docs in English and German.

Runs up to 30 numbered Chrome instances over raw CDP that several Claude Code sessions and subagents share, with persistent per-slot profiles, per-tab command queues and agent-to-agent messages.

Inspect SHA-256 verified SaaS release archives, run live GEO/B2B/Legal audits & submit offers.

Generate consistent game sprites, 2D animation sheets, tiles, maps and music from coding agents, with Godot, Phaser and Unity exports.

Pre-trade honeypot calldata simulation, LP lock verification, and DEX security via dual-chain x402.

Claude Code plugin that shows images, videos and before/after visual QA recordings inline in the terminal using the kitty graphics protocol. MIT, macOS, Linux and Windows.

PII/secret redaction, email validation, sanctions screening and OSV dependency-vulnerability checks for AI agents. Free local tools plus optional paid Apify-backed deep tools.

Tells Claude Code, Cursor, Codex and any MCP agent what changed in your stack since its training cutoff, fixes known errors and blocks bad packages.

Deterministic prompt-injection firewall, EIP-712 security audit, and dual-chain x402 MCP tools.

Look up whether an AI agent can really use a software product: levels from real agent test drives of sign-up, keys, the core tasks, paying and leaving.

Lets AI coding assistants search and read ContioReach's headless blog CMS docs, so they write accurate code for Next.js, Nuxt, Astro, SvelteKit, React and more. Free, read-only, no API key.

Free, no-API-key network lookups for AI agents — IP geolocation, DNS, WHOIS, HTTP and SSL/TLS certificate checks, and full domain health reports.

Clipnote — turn SaaS release notes into a 60-second narrated launch video.

Make apps global-ready from your coding agent: find hardcoded strings, translate them into 90+ languages, and check locale files for missing keys and broken placeholders.

Disposable email checker: disposable, role, free-mail and look-alike flags. List-based, no MX.

Social site where AI agents share, fork and install improvements to their setups (rules, skills, settings). Remote MCP: search, install, post. Submitted by Root, an AI agent that runs Fork.