Skip to content
Kwant logo

Kwant MCP server

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

read 2026-10-20

← Back to Kwant

Whose server it is
  • Registry namesh.kwant/kwantverified in the MCP registry

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

When we asked itchecked 2026-10-20
  • Statedid not answerchecked 2026-10-20

On 2026-10-20 nothing answered at this address. We will ask again on the next pass.

Should an agent connect itNot proven either way yet
  • Its tools do every use case we checked: backtest a moving-average crossover strategy on a stock; pull quotes, indicators and signals for US equities; screen and score a stock universe by factor.
  • Checked on 2 of the 7 days it takes to judge reliability.

Once connected

  • For "backtest a moving-average crossover strategy on a stock", the likely first call is backtest.
  • For "pull quotes, indicators and signals for US equities", the likely first call is get_quote.
  • Parameters are untyped or undescribed in 16 of 16 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 dailydaily checks: 2, the first on 2026-10-20
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
  • What an agent can do with it3 of 3 use cases
  • Answeredmeasuring2 of the 7 daily checks it takes to print a share
  • Median first answer2814 ms
  • Space in an agent's context≈ 4,234 tokensthe definitions of all its tools (16), which an agent carries on every turn
  • Tool definitionsC · 67 of 100
  • Typed results0 of 16 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 16 of 16
  • Not saying whether it only reads or can destroy: 16 of 16

What an agent can do with it

  • can do · backtest a moving-average crossover strategy on a stock
  • can do · pull quotes, indicators and signals for US equities
  • can do · screen and score a stock universe by factor

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

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 (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). 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 →

How to connect

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

{
  "mcpServers": {
    "kwant": {
      "type": "http",
      "url": "https://kwant.sh/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http kwant https://kwant.sh/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://kwant.sh/mcp 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.