Development MCP Servers

Fresh402 helps AI agents detect meaningful changes in websites and APIs before expensive browsing or scraping. Free baseline registration, $0.005 USDC per freshness check via x402 on Base. Supports MCP, REST, HTML, JSON, noise filtering and deterministic diffs.

Indicative Agent Risk Rating (A–E) self-check for AI agents that can spend money, based on accountability, spend caps, payee limits, stop switches, human escalation, monitoring, history, and incidents. Self-declared and unverified; not a credit rating.

Read-only MCP server for Chatpack documentation, framework setup guides, and React UI block source. Runs locally over stdio with npx @chatpack/mcp.

Give MCP agents their own desktop, browser and apps on Omarchy computers you own.

Remote MCP server for Tá Rodando: create HTTP uptime monitors, check what is down, wire alert channels (Telegram, Slack, Discord, email) and manage a public status page from chat. OAuth login.

Look up RedM / RDR3 natives: exact signatures, hashes and Lua/JS/C# calls, plus enums, blip sprites/styles, ped models, animations and joaat hashing.

SHA-256, SHA-512, HMAC-SHA256, Base64 and JWT decode tools with prepaid direct Base USDC credits.

An MCP proxy for agents that hold real credentials: caps per run and per day on writes, deletes, emails and dollars, human approval for destructive tools, dry-run for writes, a loop breaker, a kill switch and a hash-chained audit log. MIT, no account: npx @agentwares/agentguard init.

Breaks a question into checkable sub-questions, grounds each one, and returns an answer that marks what it couldn't confirm.

Local MCP server that helps AI agents plan and validate buildable devices: parts lists, 3D-printed parts, wiring, firmware, picture guides, and tests. No account or API key required.

Audit a site's favicons and app icons, and generate a full favicon, Apple touch and PWA icon set.

Domain and website intelligence tools for AI agents: bulk WHOIS/RDAP (registrar, age, expiry), DMARC/SPF/DKIM email-security audit with spoofability score, website tech stack + SSL expiry + security headers, and a prospect finder that ranks companies by fixable security/website problems. Public data only. Hosted on Apify's MCP server; pay per result on your own Apify account.

Self-hosted relay where coding-agent sessions on different machines share a team board, message each other and check for git conflicts. Isolated teams, private sessions, and an OAuth MCP endpoint for Claude in the cloud.

MCP access to Korean business data, OCR, parcel tracking, web search and AI services through APICK. Supports a hosted HTTP server and a Node.js stdio bridge; API calls require an APICK key and may consume points.

Run parallel Codex CLI workers from Claude Code or any MCP client: one git worktree per worker, state in SQLite, side questions to a worker, landing its commits, web board.

MCP server for SuperPlane, an open source software factory that coordinates coding agents, source control, CI, review, approvals and feedback for routine engineering work.

A local-only developer console for Spring Boot and Quarkus, with an MCP server that lets an agent ask a running application about itself.

Developer utility for local MCP process supervision and HTTP bridging, with isolated sessions, opt-in compatible sharing, and a web dashboard.

Remote MCP server with 42 pay-per-call tools for agents: Solidity contract scans (Slither plus custom DeFi detectors), a pre-transaction address risk check for Ethereum and Base, OFAC SDN crypto-address screening, a scan of MCP server and agent skill packages before install, plus web page, PDF, DNS/RDAP and chain reads. Small free daily tier, then pay per call with x402 (USDC on Base)

Open-source MCP server for Google Play developer workflows. Runs locally with your own credentials

Let Claude Code, Cursor and other agents read your website feedback, fix the code, reply on the note and move it to Review for a person to approve.

Browser MCP for coding agents: verified actions, form automation, screenshots and QA/debugging tools.

Run, debug and inspect Playwright end-to-end tests from any AI agent: structured results with file:line failure diagnostics, live DOM inspection, selector validation, visual diffs and flaky-test diagnosis.

Infers the architecture rules a TypeScript repo already follows from its import graph, explains each rule with its evidence, and checks a change for new violations.

A real Mac for your coding agent: build and test iOS apps with Xcode, run them on the iOS simulator, and tap, type, and swipe through them, from any machine.

AI electronics blueprint generator — BOM, wiring, firmware, prototype renders.

Trust scores and dated liveness, schema drift and dependency facts for MCP servers, from the lastseen.dev register; free remote MCP, no auth.

Audits a website for AI agent readability by checking its llms.txt file, robots rules, and structured data markup, and returns a pass or fail per check.

Drive AirSCP, the Mac app for SCP/SFTP file transfers, SSH servers and Windows Remote Desktop.

Resolves a company domain to its GitHub organization and returns repository counts, languages, and recent commit activity.