AI & Machine Learning

getjson.de

An MCP server that extracts structured JSON from documents using AI models and manages extraction jobs and credit balances.

1 endpoint12 known toolsFirst detected October 17, 2026Last detected October 17, 2026

ENDPOINT 1

https://getjson.de/mcp

No auth detected

MCP server metadata

Name: getJSON
Version: 1.0.0

Capabilities

tools

Server instructions

getJSON turns documents (PDF, images, text, CSV, X posts, web links) into structured JSON. Without an API key, call trial (proof-of-work, no email) and configure the returned key as 'Authorization: Bearer gj_...', then call extract. Give the returned claim_url to the user so they can keep the key. For a full account, call signup, then verify. To only read a link's metadata (title, description, favicon, Open Graph) or an X post, call fetch_link_metadata or fetch_x_post: no AI, 0 credits.

Known tools 12

extract

Extract structured JSON from a document.

Inferred read-only

get_job

Get the status and the extracted JSON result of a job.

Inferred read-only

list_jobs

List recent extraction jobs, newest first.

Inferred read-only

get_balance

Get the remaining credit balance.

Inferred read-only

list_models

List extraction modes and AI models with their credit cost.

Inferred read-only

list_schemas

List saved output schemas.

Inferred read-only

create_schema

Save a JSON Schema for reuse with extract's schema_id.

Inferred read-only

fetch_x_post

Fetch an X (Twitter) post as normalised JSON without AI extraction.

Potential side effects

fetch_link_metadata

Fetch the metadata of any public web page without AI extraction: title, description, canonical URL, site name, language, author, dates, keywords, preview image, favicon and icons, every Open Graph and Twitter card tag, JSON-LD and the oEmbed URL.

Inferred read-only

trial

No API key yet?

Inferred read-only

signup

Want a full account instead of a trial?

Inferred read-only

verify

Exchange the emailed 6-digit code for a new API key (gj_...).

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.getjson]
url = "https://getjson.de/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "getjson": {
      "type": "http",
      "url": "https://getjson.de/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: getjson
Remote MCP URL: https://getjson.de/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": {
    "getjson": {
      "url": "https://getjson.de/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "getjson": {
      "type": "http",
      "url": "https://getjson.de/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "getjson",
  "transport": "streamable-http",
  "url": "https://getjson.de/mcp"
}

MCP Inspector

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

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

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