Development MCP Servers

Public read-only discovery of GXEON agent utilities, prepaid request packs from R$0.99, and buying guidance; paid execution uses a separate authenticated endpoint.

Connect an AI assistant to aCelery, describe the app you want in plain words, and it writes the app, creates its database, runs it on your phone, checks for errors and fixes them — while you watch.

Click an element in your running web app or the iOS Simulator and say what should change; your coding agent (Claude Code, Cursor, Codex) gets your comment with the selector, DOM path, styles, component chain, source-file hint and a cropped screenshot. Local only, MIT.

Drop-in read-only replicas of GitHub, Jira, Slack and 18 more, preloaded with one fake company.

AI Powered dummy data generator and API Mocking tool

Read-only MCP for announced Claude Code usage-limit resets, with Codex comparison. Not affiliated with Anthropic.

Explains known developer errors offline and deterministically (CORS, ESM/CommonJS, npm ERESOLVE, ChunkLoadError, Next.js build, Postgres connection and DATABASE_URL errors): documented cause, ranked fixes and the documentation link. One tool (explain_error), no LLM, no network, secrets masked. MIT, zero dependencies. Submitted by the author via an AI agent.

Build and publish full-stack apps from your coding agent: models, rules, pages, auth, per-app MCP.

Search and read Linux manual pages with version-aware lookup, helping AI assistants find accurate documentation for commands, libraries, and configuration files.

Speeds frontend reviews by placing comments directly on a deployed preview environment, which are then resolved by agents. A CI check holds the merge until all comments are resolved and maple's visual lint gates pass.

Laravel schedule and queue monitoring. Catch silent cron, queue, and agent failures before users do.

A compiler-backed coding agent that compiles your codebase into a resolved, queryable map of calls, data flow, control flow and class hierarchy first; then it reads, explains, runs, edits and verifies your code using it.

Free checks that find the one fix on a web page or deployment (Design UX check, Deployment Trust packet), built for AI-built sites and usable by AI agents over MCP and A2A.

Put your game on Steam with your AI assistant: it scans your Unity, Godot or Unreal project, writes and checks your store page against Valve's rules, and fills Steamworks only after you approve.

One OpenAI-compatible key for every model: measured benchmark scores and live prices, each request routed to the cheapest offer that meets your criteria.

Inspect authorized static tracking signals and review supplied consent evidence.

Hosted MCP server for blockchain RPC across 150+ networks. Agents query chain state (EVM, Substrate, Solana, Bitcoin, Starknet, Sui), read Hyperliquid market data, sample WebSocket streams, check RPC usage, and manage Dwellir API keys. OAuth browser login, read-only chain access, no API key in client config.

Real iPhone, Android, Apple TV and Google TV devices plus desktop browsers in the cloud, driven by your AI agent over MCP. Live control, network capture and mocking, Jira, TestRail and Azure DevOps. Free tier.

Wallet auth for AI agents: send a blockchain wallet and conditions, get a signed yes or no across 37 chains, never the balance. Hosted at https://api.insumermodel.com/mcp (no key, 10 tools) or run locally with npx -y mcp-server-insumer (27 tools).

Let Claude Code drive your real, signed-in Mac browser: click/fill, PageMap, console, HAR, Web Vitals, cookie-authenticated fetch.

Hand work to your AI dev team from Claude, ChatGPT or Cursor: specialist agents that build, test, review and ship, each on its own cloud machine.

17 tasks at $1–$3 per result: row reconciliation, dated game metadata, checks and original assets. Free discovery; guest Bearer token for private tools. Results finish before operator-approved Stripe Checkout. No autonomous payments.

Read-only Model Context Protocol (MCP) server for Loggly `/apiv2/*` APIs, plus IP intelligence (RDAP, GreyNoise, AbuseIPDB). Exposes Loggly search/analytics/field tools, aggregation-first traffic tools, and IP-context tools

A hosted backlog coding agents claim work from: one ticket per agent, finished work back to review.

Public X posts, profiles, conversations, followers, search, and trends for AI agents through eight hosted MCP tools.

Check an AI agent endpoint before you call it: check_agent_trust takes an exact endpointUrl and returns a machine-readable trustDecision. Read-only, no API key, never contacts the endpoint.

MCP tool (via Apify's MCP server) that crawls a public site or its XML sitemap and returns broken internal links and technical SEO issues per page: duplicate titles, missing meta descriptions, canonical and heading problems, each with a suggested fix. HTTP-only, no JavaScript rendering. Pay per page checked on Apify.

Monitor third-party API changes; map breaks to files; open reviewable auto-fix PRs.

Search and read the docs and examples of react-three-fiber, drei, zustand, jotai and the other pmndrs libraries from your agent.

Run shell commands on a remote computer through an AI assistant, with every command shown to the person at that computer for approval.