
Claude Desktop
Desktop · Freemium · Proprietary
Anthropic's official Claude AI desktop application. Supports MCP servers to extend functionality.

cyanheads/toolkit-mcp-server
Generate random IDs, QR codes, and hashes, encode and decode values, and geolocate IPs, plus gated network and system diagnostics, via MCP. STDIO or Streamable HTTP.
bunx @cyanheads/toolkit-mcp-server@latest{
"mcpServers": {
"toolkit-mcp-server": {
"command": "bunx",
"args": [
"@cyanheads/toolkit-mcp-server@latest"
],
"env": {
"MCP_TRANSPORT_TYPE": "<MCP_TRANSPORT_TYPE>",
"MCP_LOG_LEVEL": "<MCP_LOG_LEVEL>"
}
}
}
}MCP_TRANSPORT_TYPEMCP_LOG_LEVELLocal, pure-compute core — hashing, ID minting, QR encoding, and value encode/decode run entirely in-process via node:crypto and the qrcode library; no upstream calls
Fail-closed gating — the two host-probing tools (toolkit_check_network, toolkit_check_system) are absent from tools/list unless explicitly enabled, so a hosted instance exposes no SSRF or…
Two-tier network gate — even with diagnostics enabled, private/reserved/loopback/link-local targets (including the cloud-metadata endpoint) stay blocked until a second flag permits them
Bounded inputs — QR data capped at 2953 bytes, rendered PNGs capped at 2048 px per side, ID batches capped at 1000; CSPRNG-backed primitives with constant-time hash comparison via timingSafeEqual
Provenance — geolocation echoes resolvedIp (the IP actually located) and source (the answering provider); absent upstream fields are reported as unknown, never invented
Response shaping — provider-supplied strings (org, isp, as) are length-bounded and stripped of control characters before they reach the response, so untrusted registry text can't flood or format a…
Discriminated output contracts — operation, format, mode, and what fields echo back exactly what ran, with only the branch-relevant fields populated per call; an unreachable host in…
Typed failure reasons — decode, hashing, QR, geolocation, and network failures each carry a structured reason plus a next-step recovery hint (e.g. decode_not_utf8, expected_malformed,…
toolkit_hash_value
Generate a cryptographic digest (sha256/sha384/sha512/sha1/md5) as hex, base64, or SRI, or constant-time-compare a value against an expected digest.
toolkit_generate_id
Mint cryptographically-random identifiers — UUIDv4, UUIDv7, or ULID — singly or in batches up to 1000.
toolkit_generate_qr
Encode text or a URL into a QR code as SVG markup, base64 PNG, or a terminal-renderable string.
toolkit_encode_value
Encode or decode a value across base64, base64url, hex, or URL percent-encoding, in either direction.
toolkit_geolocate_ip
Resolve a public IP or hostname to geographic and network metadata — country, city, coordinates, ASN, timezone.
toolkit_check_network
Gated, off by default.
toolkit_check_system
Gated, off by default.
operation
generate (a digest) or compare (timing-safe check via timingSafeEqual); omitted, it compares when expected is sent and generates otherwise. generate sent…
digestEncoding
sets the generated digest's form: hex (lowercase, default), base64, or sri (sha512-<base64>, the npm lockfile integrity and Subresource Integrity form —…
expected
is accepted as hex, base64, or SRI, recognized by its shape at the algorithm's digest length, so a published checksum is pasted as-is.
inputEncoding
reads value as utf8 (default), hex, or base64, so binary blobs skip a decode round-trip
count
mints a batch up to 1000 in one call; the returned ids array always holds exactly count values
uuid_v7
and ulid batches are monotonic — strictly increasing even within the same millisecond — so ids stays in sorted creation order.
format
svg (inline markup), png_base64 (raster bytes with mimeType and byteLength), or terminal (plain Unicode half-blocks with no escape codes, fenced in content[])
terminal
is drawn for a dark background: light modules, quiet zone included, are blocks and dark modules are spaces
errorCorrection
(L/M/Q/H) trades data capacity for damage tolerance; margin sets the quiet-zone width in modules for every format; scale sets pixels per module for svg (its…
png_base64
also arrives as an MCP image content block, so a client reading content[] can render the code without decoding structuredContent
encoding
base64, base64url (URL-safe alphabet), hex, or url (percent-encoding)
outputEncoding
(decode only) returns the recovered bytes as utf8 text (when omitted), hex, or base64 — lossless for binary data, and a direct transcode between encodings (a…
proxy
hosting, and mobile flag when the address is a proxy/VPN/Tor exit, a datacenter network, or a mobile carrier — a true on any of them means the coordinates…
Gated
registered only when TOOLKIT_ENABLE_NET_DIAGNOSTICS=true; absent from tools/list otherwise
mode
ping (ICMP round-trip), traceroute (hop path to the target), connectivity (raw TCP connect to target on port), or public_ip (the host's own egress IP)
ping
reports sent, received, and packetLossPercent alongside the average rttMs; on macOS/BSD an IPv6 target runs ping6/traceroute6
connectivity
reports an outcome — open, refused (nothing listening), timeout (traffic dropped), or unreachable (no route) — and the connect time as rttMs when open
what
os, cpu, memory, load, or interfaces
memory
reports availableBytes (headroom for new allocations) and, when the server runs under a container memory limit, limitBytes; totalBytes, freeBytes, and…
Details on this page are taken from the project's README. Open README
Clients mentioned in this server's README:

Desktop · Freemium · Proprietary
Anthropic's official Claude AI desktop application. Supports MCP servers to extend functionality.

Desktop · Freemium · Proprietary
The first agentic IDE. The Cursor editor truly merges how developers and AI work together, delivering a magical coding experience.
Desktop · Freemium · MIT
VS Code integrates MCP with GitHub Copilot through agent mode, allowing direct interaction with MCP-provided tools in your agentic coding workflow. Configure servers in Claude Desktop, workspace, or user settings, with guided MCP installation and secure handling of secrets in input variables to avoid leaking hardcoded keys.

microsoft/playwright
Playwright is a framework for web automation and testing. It drives Chromium, Firefox, and WebKit with a single API — in your tests, in your scripts, and as a tool for AI agents.

yamadashy/repomix
Repomix is a tool that packs a codebase into an AI-friendly format, supporting local and remote repository processing and providing code compression, security checks and multiple output formats.

bytedance/UI-TARS-desktop
TARS is ByteDance's multimodal AI agent stack, shipping two projects: Agent TARS (a CLI and Web UI agent built on MCP) and UI-TARS-desktop (a desktop GUI agent).

ahujasid/blender-mcp
formerly blender-mcp — the PyPI package is now mcp-for-blender. Existing setups keep working; no config change is required. Read more.

microsoft/playwright-mcp
A Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright. This server enables LLMs to interact with web pages through structured accessibility snapshots, bypassing the need for screenshots or visually-tuned models.

comet-ml/opik
Opik is the open-source LLM observability and evaluation platform for AI agent tracing, LLM evaluation, prompt management, and production monitoring. Built by Comet. Apache-2.0 licensed, free to self-host the full platform, with 20,000+ GitHub stars.