"Using npx to run tools directly without requiring an API" matching MCP servers:

Enables you to audit your AI agent skills by running each one against an agent that cannot see it, diffing the resulting artifacts, and grading whether each skill genuinely improves, changes nothing, or worsens the output.

Enables users to search and read verbatim passages from Terry Bouricius's book 'Democracy Without Politicians', honoring the author's request to display his actual words rather than AI summaries.

A powerful MCP server for fetching and transforming web content into various formats (HTML, JSON, Markdown, Plain Text) with ease.

Enables fetching weather forecasts for the USA via NWS API and for Israel via browser automation using Playwright.

AI-powered MCP server for file renaming, organization, and PDF splitting using natural language prompts and content analysis.

Enables AI agents to observe and control Android devices and emulators through plain ADB — taking screenshots, reading the UI hierarchy, tapping, swiping, typing, launching apps, reading logcat, and recording the screen. It also executes multi-step action sequences (tap, swipe, key, wait, text, wait-for-element) in a single tool call.

MCP server exposing a parametric policy engine for evaluating, simulating, and managing rules, with dynamic plugin support for domain-specific calculations.

Enables MCP-compatible AI applications to perform read-only npm package and project dependency safety scans, returning proceed, caution, or block verdicts without executing package code.

Provides tools for executing shell commands both synchronously and asynchronously with real-time output streaming and process management capabilities. It enables users to start background tasks, monitor progress, and manage long-running processes via Stdio or HTTP transports.

Executes SQL queries via ConnectorX and streams results to CSV or Parquet files, supporting multiple databases and optional token counting for CSV output.

Enables agents to convert Markdown into WhatsApp-compatible formatting, including styled text, lists, and tables that fit a phone's monospace width.

MCP server for personal knowledge management providing AI-powered tools for semantic search, audio transcription, image generation, and PPT creation.

Enables coding agents to inspect repository state, create and amend commits, rewrite history and manage stacked branches, worktrees, and pushes against any Git remote.

Enables MCP clients to turn Make blueprints and Zapier exports into n8n workflow files, with a per-step report of what converted, what to review, and what must be finished by hand. Also lists zaps, explains how individual steps are handled, checks workflow files for broken links or placeholders, and can upload the result to an n8n instance as an inactive workflow.

Enables coding agents to inspect chats, messages, contacts, and history, manage received files, and perform controlled sends and actions on the user's own account.

Enables coding agents to browse chats, read and search messages, look up contacts, save received files, and send messages or files through a linked account.

Provides five safe tools (calc, search, model_drift, compare_runs, grade_answer) for arithmetic, keyword search, model drift monitoring, eval comparison, and answer grading against sources, all implemented directly from the MCP spec with zero dependencies.

Enables coding agents to read Telegram chats and messages, manage mirrored chats, send messages, and save or send files on the user's own account.

Enables AI assistants to keep Rojo serving and auto-connected for Roblox Studio projects and to import FBX, OBJ, GLB, or GLTF models through Studio's own 3D Importer on Windows.

An MCP server that provides a persistent sandbox for AI coding agents to explore codebases server-side, returning only compact summaries to reduce context consumption.

Enables extraction of recipes from PDFs or JPG files and conversion into structured HTML, with interactive region selection and OCR via a web GUI.

Enables validating JSON/JSONC, getting/setting values with jq-like paths, pretty-printing or minifying, validating against Ajv JSON Schema, and diffing local files or inline strings without network calls.

Exposes local document merging, document comparison, site auditing, and release checking as standard MCP tools so Claude Desktop, Cursor, and any MCP client can invoke them directly without switching to a terminal.

Enables coding agents to run Playwright e2e tests, validate Supabase RLS policies with role-based SQL, typecheck TypeScript, execute npm scripts with parsed test output, and validate Next.js builds, all through a structured MCP interface.

Routes one brief to the right image model across 60+ (gpt-image-1.5, Ideogram 3, Recraft V4, Flux), validates the output, and fans out to iOS/Android/PWA/favicon/visionOS/Flutter bundles. Works without an API key via Pollinations, HF Inference, Stable Horde, or host-LLM inline SVG.

Turns an OpenAPI spec into a ready-to-run test harness with local LLM agents: WireMock stubs, contract tests, robustness probes, and chaos scenarios you can switch on and off in a running WireMock.

Enables AI assistants to search YouTube, fetch transcripts, and get AI summaries of videos without API keys.

An autonomous MCP server that uses the Claude Code CLI to solve coding problems without permission prompts.

Enables turning requirements documents into RFCs, functional test scenarios, and a traceability matrix, with tools to parse, diff, and verify coverage.

Enables agents to compress JPEG, PNG, and WebP images to a target byte size, split PDFs by page or ranges, and remove image metadata while processing local files without sending file bytes over the network.

Enables searching and retrieving documentation from crawled documentation sites as an MCP server, allowing coding agents to query real docs instead of relying on training data.

Exposes AWS Palace finite-element electromagnetic simulation as a set of JSON-in/JSON-out tools, so an agent can build geometry and meshes, generate and lint configs, estimate cost, guard and launch runs, and read back S-parameters, eigenmodes, Q and energy results. Results return as JSON text plus structured content over a dependency-free stdio JSON-RPC connection.

Turn natural language into 3D models in Fusion 360. 64 CAD tools including sketches, extrudes, fillets,and JIS standard parts.

Provides web automation utilities, including taking Camofox browser tab screenshots and saving them directly to local files.

A Model Context Protocol server that converts Markdown content to HTML format.

An MCP server that converts Excel and Apple Numbers files to PDF format, enabling AI assistants like Claude to perform file conversion directly through conversation.

Enables an AI assistant to control a Hyprland/Wayland Linux desktop through 178 tools across 30 categories, covering audio, window and workspace management, themes and wallpapers, screenshots and recording, Bluetooth, media playback, notifications, system stats, and chained multi-step macros. Everything runs locally via hyprctl, PipeWire, and standard Linux utilities, with no cloud dependencies.

Enables an MCP-compatible AI client to author Blender Python, run Blender, import the resulting asset into Unity with materials and a prefab, place it in the active scene, verify the import, and revise the same asset using stable IDs.

MCP server that fetches API responses and generates ready-to-use model/POJO classes in your configured language, with built-in login/refresh token handling.

Enables AI agents to debug and repair historical Solaris/SPARC systems by providing an out-of-band control plane with guest DTrace, QEMU monitor access, SPARC-aware GDB, host eBPF/perf tracing, and an immutable evidence ledger for cross-layer hypothesis testing.