← Registry

General Tools

unscramblekey.com

Provides word puzzle tools for unscrambling letters, finding anagrams, solving Wordle, fitting crossword patterns, and listing dictionary words.

1 endpoint6 known toolsFirst detected November 12, 2026Last detected November 12, 2026

ENDPOINT 1

https://unscramblekey.com/api/mcp

No auth detected

MCP server metadata

Name
unscramblekey-com
Version
1.0.0
Capabilities
toolsresourcesextensions.io.modelcontextprotocol/ui.mimeTypes
Server instructions

UnscrambleKey is an English word unscrambler and puzzle assistant. unscramble_letters lists every word a rack of letters can make, grouped by length with common words first; _ or ? is a blank tile. find_anagrams lists the words that use all the letters. solve_wordle takes each guess with the colour of every letter (green, yellow, gray) and returns the words still possible and the best next guesses. solve_pattern fits a pattern such as _A_E_. word_list lists words by length, start, end or contained letters. score_word checks one word and gives its Scrabble and Words With Friends points. Scoring is scrabble unless the user names Words With Friends (wwf). Report only the words and numbers the tools return; do not add words from memory. Words in bold or marked common are everyday words; the rest are valid but rare. A result lists the first words; a cursor in it returns more, so request it only when the user asks for more words, with the same arguments. Offensive words are hidden; set include_offensive only when the user asks for them. Give the returned link so the user can continue on unscramblekey.com with the same input.

Known tools 6

unscramble_letters

Finds every dictionary word that can be made from a rack of letters, with optional blank tiles and start, end, contains and length filters.

Inferred read-only
find_anagrams

Finds the words that use all of the given letters exactly once, with blank tiles allowed.

Inferred read-only
solve_wordle

Solves a Wordle position from the guesses made and the colour of each letter.

Inferred read-only
solve_pattern

Finds the words that fit a crossword-style pattern such as _A_E_, where _ is an unknown letter, optionally limited to words whose unknown positions can be filled from the available letters.

Inferred read-only
word_list

Lists the dictionary words of a given length, or that start with, end with or contain given letters, in popularity order with common words first.

Inferred read-only
score_word

Checks a word against the dictionary and scores it.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.unscramblekey-com]
url = "https://unscramblekey.com/api/mcp"
enabled = true
Claude Code

.mcp.json

{
  "mcpServers": {
    "unscramblekey-com": {
      "type": "http",
      "url": "https://unscramblekey.com/api/mcp"
    }
  }
}
Claude Desktop

Settings → Connectors → Add custom connector

Name: unscramblekey-com
Remote MCP URL: https://unscramblekey.com/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "unscramblekey-com": {
      "url": "https://unscramblekey.com/api/mcp"
    }
  }
}
Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code
{
  "servers": {
    "unscramblekey-com": {
      "type": "http",
      "url": "https://unscramblekey.com/api/mcp"
    }
  }
}
Generic MCP

Client-specific MCP configuration

{
  "name": "unscramblekey-com",
  "transport": "streamable-http",
  "url": "https://unscramblekey.com/api/mcp"
}
MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for unscramblekey.com was fetched 2026-11-13T19:14:00.944Z.

Trust status Neutral

unscramblekey.com is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.