Skip to content
ScoreFluent logo

ScoreFluent MCP server

Which MCP servers are ScoreFluent's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-11-05

← Back to ScoreFluent

Whose server it is
  • Registry namecom.scorefluent/scorefluentverified in the MCP registry

The registry name sits under scorefluent.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-11-16
  • Stateanswerschecked 2026-11-16
  • Tools10
  • Server calls itselfscorefluent
  • Server version0.0.1
  • Protocol version2025-06-18

On 2026-11-16 the server answered initialize and tools/list without a sign-in. First answer in 249 ms.

Should an agent connect it

Not proven either way yet
  • Whether its tools do the product's job has not been judged yet.

Once connected

  • Parameters are untyped or undescribed in 3 of 10 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured daily

daily checks: 16, the first on 2026-11-05
  • INITE MCP score89 of 100answers 100 · definitions 95 · context 100 · connects 100 · lowered until one of our scenario runs passes
  • Answered100% of days16 of 16 daily checks
  • Median first answer263 ms
  • Space in an agent's context≈ 2,493 tokensthe definitions of all its tools (10), which an agent carries on every turn
  • Tool definitionsA · 95 of 100
  • Typed results10 of 10 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 3 of 10

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers10 tools · read 2026-11-20
  • exam_overview

    Use this when a student asks what an English exam contains, how a task type is scored, how long a section is, what the score scale is, or where to start preparing. Returns the exam sections with task types, question counts and time…

  • question_type_skills

    Use this when a student asks what a question type tests, which skills it needs, how to practise for it, or what mistakes people make on it (for example "what does IELTS Writing Task 2 test?" or "what skills does PTE Read Aloud need?")…

  • practice_question

    Use this when a student wants a real practice question for an English exam task, for example "give me a PTE Write Essay prompt" or "a TOEFL listening question". Returns one random published question (prompt, passage, options, word limit…

  • score_writingchanges data

    Use this when a student pastes an essay, email or summary for a PTE Academic, LanguageCert Academic, TOEFL iBT, MET, CELPIP, IELTS Academic, IELTS General Training, Cambridge C1 Advanced, PTE Core or Oxford Test of English (beta) writing…

  • start_diagnostic

    Use when a student does not know which exam to take (Exam Match) or knows the exam and wants a starting baseline (placement). Returns a link to the test on the site; do not try to run the test in chat.

  • check_read_aloud

    Use when a student has read a Read Aloud passage aloud and you have the transcript of what they said (for example from voice mode) but no audio. Compares the words said against the passage: omitted, inserted and substituted words and a…

  • speak_practice

    Use whenever a student asks for a PTE Read Aloud or Describe Image to practise, even without mentioning recording or scores. Lets them practise the speaking task and get pronunciation and fluency scores. Shows a recorder card; desktop web…

  • recorder_startchanges data

    Reserve a recording slot and get an upload URL. Called by the speaking recorder card only.

  • recorder_confirmchanges data

    Confirm the upload and queue it for scoring. Called by the speaking recorder card only.

  • recorder_result

    Read the scoring status and scores for a recording. Called by the speaking recorder card only.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "scorefluent": {
      "type": "http",
      "url": "https://api.scorefluent.com/mcp/copilot"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http scorefluent https://api.scorefluent.com/mcp/copilot

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://api.scorefluent.com/mcp/copilot as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.