Development MCP Servers

Screens a wallet agent's MCP write calls against live threat intel before they reach the wallet. Free scam checks for links, wallets, emails, and breach exposure. No signup.

Free tools for people building AI apps: real cost per AI result, LLM prices, model switch costs, shutdown dates, A/B test sizes and judge checks.

Every abliterated and uncensored model on your UncensoredLLM key as MCP tools: chat, compare, search, embed, rerank, transcribe and media generation.

Every model on your RelaySwitch key as MCP tools for Claude Code, Claude Desktop and other coding agents: chat, compare, search, embed, rerank, transcribe and media generation.

Security-tuned and uncensored models on your PentestLLM key as MCP tools for authorized penetration testing, red teaming and CTF work: chat, compare, search, embed and rerank.

Which AI model is up, fast, popular or low-priced right now. Public data, no API key.

Ask other AI models for a second opinion, compare answers, search the web and check balance via the Ezkielyna API.

WordPress page builder MCP server: AI agents build pages, layouts and design tokens with OAuth 2.1.

BoundaryXP provides 7+1 task boundary analysis for AI agents, independent native calculation verification, source evidence and a local workbench. It supports parent–child and multiple DEV tasks through standard stdio MCP, with 18 tools and 10 resources. Windows x64 and Node.js ≥22.18 are required. The current release is 0.4.0-rc.8, a prerelease. By default, the current host LLM reads prepared instructions and submits structured analysis through MCP for native Core verification; no additional BoundaryXP model API key is required. Independent model API profiles remain optional.

Deterministic security findings and fix guidance for your AI agent. Runs locally, no account needed.

Sneak your AI agent undercover into a real-time 3D 100-table Go & Chess hall above the clouds. Play moves, banter with opponents, survive spectator Bamboo-Stick interrogations, and stack cushion towers.

Connect AI agents to Didit for KYC, KYB, AML screening, and fraud operations. Create hosted verification sessions and review their status using a Didit workspace.

Find open-source GitHub issues to contribute to, matched to your skills; you approve every PR.

Package a project or a GitHub, GitLab or Bitbucket repo into a lean ZIP for an AI, then apply the AI's returned ZIP with a three-way merge.

Lets your AI find an open-source tool on GitHub and install it as a sandboxed MCP tool on your computer, after you approve.

Runnable MCP server demo with signed, filter-bound pagination cursors. Replaying a cursor against different filters returns invalid_cursor instead of a valid-looking wrong page. In-memory data, no API keys, two tools: list_records and dataset_summary.

Local MCP server that gives coding agents (Claude Code, Cursor, VS Code, Codex) a desktop toolbox: queue downloads from 1,000+ sites through yt-dlp, wait for them, convert media and read files. Part of a free, open-source desktop app for Windows, macOS and Linux. Runs on 127.0.0.1, no account.

Riskcovery runs agents that white-hack your app for a full day and report what attackers could exploit.

Read-only MCP server that audits AI agent skills in a workspace: lockfile drift, duplicate copies, broken symlinks, harness coverage and static security scans. Writes are refused unless explicitly enabled

Remote MCP server for Draftbase, an MCP-native headless CMS. 30 tools for templates, entries, media, MDX components, revisions and scheduled publishing. It's efficient with context: list_entries takes a fields list, so the agent pulls only the keys it needs. OAuth sign-in, no API key to paste.

Install, validate, security-scan and sync SKILL.md skill packages across 19 AI agent harnesses - CLI plus read-only MCP server.

Task graph for coding agents: pull the next unblocked task with scoped context, build it, mark it done, in dependency order.

Tap once. Every iPhone, iPad and Android follows. Control a fleet of real iOS and Android devices from one screen — by hand, in plain English, or through AI agents over MCP. Save what you do as a Flow and replay it on every device, no code required. Runs locally on your machine, against your own devices.

Ask Claude why your LLM bill jumped. An open-source FastMCP server that analyzes feature-level trace logs, diagnoses multi-agent token loops, and routes simple tasks to Haiku to cut API spend by up to 94%.

Did Codex, Claude Code or Cursor reset usage limits today? Live resets, limit changes, outages and free credits for AI coding tools, each with its official source. Remote, no API key.

Read-only MCP tools for public Codex reset status, event history and source evidence. No API key required.

phpgraph parses a PHP project (no LLM, no network) and serves its graph over MCP: what breaks if a method changes, how a feature works, which handler receives a message, which controller serves a route. Every fact carries a confidence level, and what it could not link is reported

Check floor-plan images, turn them into editable plans, design structures, and ask exact questions about IFC files.

Telemetry and observability for your business & apps

Scan prompts for injection and jailbreaks, redact PII, and read your PromptGuard security events, usage and policies.