DataLedger MCP Server

Ask your AI agent to find any UK company

Two lines of config and Claude, ChatGPT or Cursor can query live UK company financials, ownership records and corporate structures on your behalf. No coding required.

25 free credits on signup · No card required


A real conversation, powered by DataLedger

This is an actual Claude session with the DataLedger MCP server connected. The user typed a plain English question. Claude searched 71 companies, pulled financial data on the most promising candidates in parallel, and returned a structured acquisition shortlist.

Claude + DataLedger MCP
Find me companies in the Salford area that would be a good acquisition target. I have £1m to spend max, want a growing service-based company with individual owners.
Let me search for companies in the Salford area (M5 to M7) with individual owners and signs of growth…
Called dl-mcp → company_search (postcode: M5, M6, M7)
Good, 71 results. Let me filter for service-based SIC codes and pull detailed financials on the most promising candidates in parallel…
Called dl-mcp ×10 → company_profile + financials + psc

Salford Area: Acquisition Shortlist

Criteria: Active · individual owners only · >5% asset growth · equity <£1m · non-dormant · service SIC codes

Tier 1: Strong targets
Nexbridge IT Solutions Ltd
IT Consultancy + Management Consulting
Equity +52% YoY Assets +45% YoY D/E 0.11 2 staff
Dual-service model, strong cash accumulation, virtually debt-free. Couple-owned (50/50) so a clean negotiation. Best growth profile of the group.
Meridian Project Systems Ltd
Engineering Consultancy
Equity +49% YoY Assets +32% YoY D/E 0.05 Solo consultant
Established 21 years, essentially debt-free. Owner born 1965, possible succession angle. Key-person risk to factor into price.
Castleton Accountancy Ltd
Accounting & Auditing
Turnover £232k (+6%) Profit 3× growth Equity +111% YoY D/E 0.20
Only company in this set with visible P&L. At 1 to 1.5x revenue multiple (around £230k to £350k) this is very affordable and leaves room to scale. Couple-owned. Recurring client base is the prize.
Tier 2: Worth investigating
Irongate IT Consultancy Ltd
IT Consultancy
Equity +31% YoY D/E 0.17 Est. 2011 1 staff
15 years old, good growth. Single owner so key-person risk, but accumulated equity suggests consistent profitable trading. No P&L disclosure.
Brindley & Holt Solicitors Ltd
Solicitors
Equity +76% YoY 9 staff Revenue needed
Largest workforce in this set. Low equity is normal for solicitors (£2.1m on balance sheet is client funds in trust). Goodwill value based on billings is the acquisition metric here. Obtain revenue figure before approach.

Eliminated

Helixnode Ltd - negative equity (-£5k, insolvent)  ·  Veritas Digital Ltd - negative equity (-£6k)  ·  Greystone Apps Ltd - equity only £8k, effectively micro  ·  Premier Advisory Services - owners born 1953/54, 0 staff, only 7 to 8% growth

Recommended next steps

Castleton Accountancy and Nexbridge are the cleanest targets: couple-owned, growing, service-based, healthy balance sheets, well within budget. For Brindley & Holt, obtain revenue/billings before valuing. For Meridian and Irongate, model key-person dependency carefully and build retention terms into any deal.


MCP lets AI tools use live data instead of guessing

MCP (Model Context Protocol) is an open standard that lets AI assistants like Claude connect to external data sources in a controlled way. Even AI tools with live web search cannot reliably retrieve structured financial records, PSC ownership data or corporate group structures. The DataLedger MCP server gives your AI direct, structured access to that data so it can reason over real numbers rather than approximations.

Search companies naturally

Ask in plain English: by sector, location, financials, employee count, growth rate or owner type.

Retrieve financial profiles

Balance sheet totals, year-on-year growth, ratios and P&L where disclosed in filings.

Inspect PSC ownership

Individual and corporate PSC records, ownership bands, owner age and voting rights.

Map ownership chains

Follow parent companies upward and subsidiaries downward from any single company.

Works with your AI tools

Claude Desktop, Claude Code, Cursor, Windsurf, n8n, Zapier and any other MCP-compatible client or agent framework. ChatGPT users can connect via a custom GPT with Actions.

Read-only and credit-tracked

The server only retrieves data. Nothing is written or changed. All usage maps to API credits so there are no surprises.


Connected in under five minutes

Pick your AI tool below. You need a DataLedger API key for both routes - grab one free at hub.dataledger.uk.

Works with Claude Desktop, Claude Code, Cursor, Windsurf, n8n, Zapier and any other MCP-compatible client. No code required - choose a connection method and paste the config into your client settings.

Hosted (recommended)

Connect to our hosted server

The DataLedger MCP server runs on our infrastructure at api.dataledger.uk/mcp. Add the URL and your API key to your client config and you are ready.

  • Nothing to install on your machine
  • Server updates happen on our side, not yours
  • Two config options: native HTTP (recent clients) or mcp-remote bridge (wider compatibility)
Self-hosted (advanced)

Run your own instance

Clone the repo from GitHub and run the server yourself. Gives you direct access to the source so you can adjust the system prompt and control exactly what the server requests.

  • Tune the system prompt to shape what gets returned and how
  • Tighter control over LLM token usage and DataLedger credit spend
  • Node.js required - see the GitHub repo for setup
1
Create a free account and get your API key
Sign up at hub.dataledger.uk. You get 25 free credits instantly. Copy your API key from the dashboard.
2
Add DataLedger to your AI client
Open your MCP client settings and add a new MCP server. The exact steps vary by client; see the config options below.
3
Ask your AI to start researching companies
Type a plain English question. Your AI will call DataLedger, interpret the results, and come back with a structured answer, just like the example above.

Option 1: Native HTTP (recommended)

Supported in any MCP client that handles native HTTP connections, including recent versions of Claude Desktop, Claude Code and Cursor. No install required - just add the server URL and your API key to your client config.

JSON / claude_desktop_config.json or .claude.json
{
  "mcpServers": {
    "dataledger": {
      "type": "http",
      "url": "https://api.dataledger.uk/mcp",
      "headers": {
        "x-api-key": "YOUR_API_KEY"
      }
    }
  }
}

Option 2: Via mcp-remote (fallback for older clients)

If your client does not yet support native HTTP MCP connections, use mcp-remote as a bridge. Node.js must be installed. This works across the widest range of current client versions.

JSON / claude_desktop_config.json or .claude.json
{
  "mcpServers": {
    "dataledger": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://api.dataledger.uk/mcp",
        "--header",
        "x-api-key:YOUR_API_KEY"
      ]
    }
  }
}

Option 3: Claude Code CLI shortcut

Claude Code has a CLI command that writes the config for you. Other clients with their own CLI tools will have an equivalent. This writes to ~/.claude.json and picks it up on next start.

Shell / Claude Code CLI
claude mcp add --transport http dl-mcp \
  https://api.dataledger.uk/mcp \
  --header "x-api-key: YOUR_API_KEY"

# Confirms: Added HTTP MCP server dl-mcp
# URL: https://api.dataledger.uk/mcp
# File modified: ~/.claude.json

ChatGPT does not currently support MCP connections without OAuth. The practical alternative is a custom GPT with Actions, which calls the DataLedger API directly using your API key. The result is similar: a GPT that can search UK companies, retrieve financials and work with Companies House data in conversation.

1
Create a new custom GPT
In ChatGPT, go to Explore GPTs and select Create. Give it a name and add your instructions using the example below.
2
Add an Action and choose API key authentication
Go to Actions, select Create new action, then choose API key as the authentication type and Bearer as the method. Paste your DataLedger API key.
3
Paste the DataLedger OpenAPI schema
Add the schema below to the Action. This tells ChatGPT which DataLedger endpoints it can call and what parameters they accept.
4
Test and save
Use the GPT preview to run a company search. Once it returns results, save the GPT and start using it.

Suggested GPT instructions

Paste this into the instructions field when creating your GPT. Adjust the tone and scope to suit your use case.

GPT instructions
You are a UK company data assistant powered by DataLedger.

Use the DataLedger API when the user asks for UK company searches, company profiles,
financial data, ownership data, SIC filtering, location filtering, or Companies
House-derived structured data.

Explain clearly when data is based on electronic Companies House filings. Do not imply
that turnover or profit and loss data is available for every company. Where profit and
loss data is unavailable, explain that many UK SMEs do not disclose full P&L figures.

When searching, prefer structured filters such as SIC code, local authority, company
status, incorporation date, assets, liabilities, equity, employees, and available
P&L figures.

Return concise summaries first, then structured tables where helpful.

OpenAPI schema

The full OpenAPI schema for the DataLedger API, including all available endpoints, parameters and authentication details, is on our agents and integrations page. Copy the schema from there and paste it into the Action schema field in your custom GPT.

View OpenAPI schema →
When you add a third-party Action in ChatGPT, OpenAI shows a warning that the server has not been reviewed. This is standard for any external connection. DataLedger is read-only and does not write, modify or delete any data.

Ready to connect your AI to UK company data?

Start with 25 free credits. Search live financials, ownership records and corporate structures from any MCP-compatible AI tool.