UK Business Tools - Ledgerhall
# Ledgerhall
<!-- mcp-name: io.github.paulieb89/uk-business-mcp -->
[](https://glama.ai/mcp/connectors/dev.fly.uk-business-mcp/uk-business-tools-ledgerhall)
[](https://vscode.dev/redirect/mcp/install?name=ledgerhall&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fuk-business-mcp.fly.dev%2Fmcp%22%7D)
[](https://insiders.vscode.dev/redirect/mcp/install?name=ledgerhall&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fuk-business-mcp.fly.dev%2Fmcp%22%7D&quality=insiders)
[](https://cursor.com/en/install-mcp?name=ledgerhall&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vdWstYnVzaW5lc3MtbWNwLmZseS5kZXYvbWNwIn0=)
**UK public data for AI agents.** One connection. Companies House, Land Registry, GOV.UK, courts, Parliament — your agent gets the lot.
```
https://uk-business-mcp.fly.dev/mcp
```
No API key. No account. Free and hosted.
[Set up in 30 seconds →](https://bouch.dev/products/uk-business-mcp)
---
## What you can do
Once connected, your AI agent can:
- **Check any UK company** — directors, shareholders, filing history, disqualified officers
- **Research property** — comparable sales, EPC ratings, Rightmove listings, rental yields, stamp duty
- **Search case law and legislation** — court judgments, Acts, Hansard debates, HMRC guidance
- **Query GOV.UK** — search 700k+ pages, resolve postcodes to councils, find policy documents
- **Run due diligence** — cross-reference Companies House, Charity Commission, Land Registry, Gazette insolvency, sanctions lists
## Setup
### claude.ai
1. Go to **Settings** (bottom left)
2. Click **MCP Servers**
3. Click **Add**
4. Paste: `https://uk-business-mcp.fly.dev/mcp`
### Claude Desktop
Add to `claude_desktop_config.json`:
```json
{
"mcpServers": {
"ledgerhall": {
"type": "http",
"url": "https://uk-business-mcp.fly.dev/mcp"
}
}
}
```
### Claude Code
```bash
claude mcp add --transport http ledgerhall https://uk-business-mcp.fly.dev/mcp
```
### ChatGPT
1. Go to **Settings** → **Connected apps** → **Add MCP server**
2. Paste: `https://uk-business-mcp.fly.dev/mcp`
### Cursor / other editors
Add to `.cursor/mcp.json`:
```json
{
"mcpServers": {
"ledgerhall": {
"url": "https://uk-business-mcp.fly.dev/mcp"
}
}
}
```
## What's inside
Ledgerhall is a [FastMCP](https://gofastmcp.com) proxy that bundles four specialist UK MCP servers into one endpoint. Each server covers a different part of the UK public record.
| Prefix | Server | Coverage |
|--------|--------|----------|
| `gov_` | [GOV.UK](https://github.com/paulieb89/govuk-mcp) | 700k+ GOV.UK pages, organisations, postcodes |
| `law_` | [UK Legal](https://github.com/paulieb89/uk-legal-mcp) | Case law, legislation, Hansard, HMRC guidance, OSCOLA |
| `dd_` | [UK Due Diligence](https://github.com/paulieb89/uk-due-diligence-mcp) | Companies House, Charity Commission, Land Registry, Gazette, sanctions |
| `prop_` | [UK Property](https://github.com/paulieb89/property-shared) | Land Registry, EPC, Rightmove, yields, stamp duty, planning |
## How it works
Pure proxy — no data storage, no custom tools. Every request is forwarded to the appropriate backend. Tool lists are cached for 5 minutes. Built with FastMCP `create_proxy()` and `mount()`.
## Coming next
- **UK Compliance** — FCA Register and SRA solicitor checks
- **UK Statistics** — ONS economic data and Nomis labour market profiles
New servers appear automatically. No config changes needed.
## Links
- **Product page:** [bouch.dev/products/uk-business-mcp](https://bouch.dev/products/uk-business-mcp)
- **Free AI skills:** [bouch.dev/tools](https://bouch.dev/tools)
- **All servers:** [bouch.dev/products](https://bouch.dev/products)
- **Built by:** [BOUCH](https://bouch.dev) — AI consultancy, East Midlands
## License
MIT
TDQS
Scored across 74 tools
Multiple tools overlap significantly within domains: prop_property_epc, prop_property_epc_summaries, prop_property_epc_search (deprecated but present) all handle EPC data, and gov_govuk_search vs gov_govuk_grep_content vs gov_govuk_get_section vs gov_govuk_get_content cover overlapping content retrieval. Several DD tools (dd_company_search vs prop_company_search, dd_search vs dd_fetch) duplicate functionality with unclear distinctions. The prefixes help, but within each cluster an agent will struggle to select the right tool.
Naming is inconsistent across modules: some use domain prefixes with snake_case verbs (gov_govuk_search, law_judgment_get_header), others use short opaque prefixes with no clear pattern (dd_search, prop_company_search), and one uses a different hierarchy (prop_property_epc vs prop_epc_certificate). The `law_` prefix mixes legal and DD tools inconsistently, and deprecated tools remain visible. No uniform verb_noun or consistent prefix convention.
74 tools is far beyond the well-scoped range, and the surface appears to bundle four independent domains (GOV.UK content, legal research, due diligence, property data) into one server. Many tools are near-duplicates (e.g., dd_company_search vs prop_company_search; prop_property_epc vs prop_property_epc_summaries), and the deprecated prop_property_epc_search should be removed. The count itself creates selection overhead that outweighs any benefit.
Each domain has notable gaps: GOV.UK lacks a tool to fetch full page content (only sections via govuk_get_section); property data has search and EPC tools but no explicit valuation or sale-date filtering beyond vague descriptions; law covers judgments/legislation/Hansard but lacks tools for court documents or lower-court records. The due-diligence domain is the most complete, but the overall surface has no cohesive workflow — it is four partial APIs stapled together. Deprecated tools and missing document retrieval (dd_company_filing_document requires a resource-capable client) create dead ends.