bartram-companies-house 0.3.5

UK regulatory data via MCP — Companies House (12 tools), FCA Financial Services Register (19 tools), Charity Commission (31 tools), and Environment Agency (10 tools) with cleaning, normalisation, and deduplication.

UK Companies House MCP Server

Clean, structured UK company data for AI agents — via the Model Context Protocol.

What this does

This MCP server gives agents access to 12 UK Companies House endpoints — company search, profiles, officers, beneficial ownership, filings, charges, insolvency, officer search, officer appointments, disqualification checks, advanced search, and registered addresses — with data cleaning that the raw API doesn't provide.

Companies House data has known quality issues: officer names appear in multiple formats creating ~614K duplicate records, corporate entities get filed as natural persons in PSC (beneficial ownership) data, and addresses arrive in inconsistent formats. This server fixes those problems before the data reaches your agent.

What the cleaning layer does:

• Officer deduplication — fuzzy name matching merges records like "SMITH, John David" and "John Smith" when they share the same role and company.
• PSC entity classification — flags corporate entities incorrectly filed as natural persons, using identification fields and name pattern detection.
• Address normalisation — consistent formatting, postcode standardisation, single-line formatted output across all endpoints.
• SIC code descriptions — adds human-readable industry descriptions alongside raw SIC codes.
• Nature of control interpretation — converts coded PSC control types into plain English descriptions.
• Charge status interpretation — translates charge status codes (outstanding, fully-satisfied, part-satisfied, satisfied).
• Insolvency case type interpretation — converts case type codes (CVL, compulsory-liquidation, etc.) to descriptions.
• Company number normalisation — agents can send 445790 and the server correctly zero-pads it to 00445790.
• Full pagination — officers, PSCs, filings, charges, and appointments are fetched across all pages automatically.
• Data quality annotations — every cleaned record carries _cleaning metadata so agents can see what was changed and why.

Quick start

# Install
pip install bartram-companies-house

# Set your API key (required)
export COMPANIES_HOUSE_API_KEY="your-key-here"

# Run the MCP server
bartram-companies-house

Or with uvx:

uvx bartram-companies-house

Register for a free API key at Companies House Developer Hub.

Configure with Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "companies-house": {
      "command": "uvx",
      "args": ["bartram-companies-house"],
      "env": {
        "COMPANIES_HOUSE_API_KEY": "your-key-here"
      }
    }
  }
}

Tools

Tool	Description
uk_companies_house_search	Search for UK companies by name or number
uk_companies_house_profile	Get company details — address, status, SIC codes with descriptions, key dates
uk_companies_house_officers	Get officers with name deduplication and entity classification
uk_companies_house_psc	Get beneficial owners with entity classification and nature of control interpretation
uk_companies_house_filings	Get filing history with readable descriptions, optional category filter
uk_companies_house_charges	Get charges (mortgages/security interests) with status interpretation
uk_companies_house_insolvency	Get insolvency cases with case type interpretation and practitioner details
uk_companies_house_officer_search	Search for officers by name across all companies
uk_companies_house_officer_appointments	Get all appointments for an officer across companies
uk_companies_house_disqualified_officer	Check if an officer is disqualified
uk_companies_house_advanced_search	Search companies with filters (status, SIC code, location, dates)
uk_companies_house_registered_address	Get the registered office address

See SCHEMA.md for full parameter and response documentation with examples.

Example

Ask your agent: "Who are the directors of Tesco?"

The agent calls uk_companies_house_search with query: "Tesco", gets the company number, then calls uk_companies_house_officers with that number. The response includes deduplicated officer records with cleaning metadata:

{
  "data": {
    "items": [
      {
        "name": "MURPHY, Ken",
        "officer_role": "director",
        "appointed_on": "2020-10-01",
        "nationality": "Irish",
        "_cleaning": {
          "deduplicated": true,
          "duplicate_count": 2,
          "original_names": ["MURPHY, Kenneth", "MURPHY, Ken"],
          "note": "Merged 2 records with similar names. Showing most recently appointed."
        }
      }
    ],
    "_cleaning": {
      "original_count": 45,
      "deduplicated_count": 38,
      "duplicates_removed": 7,
      "active_count": 12,
      "resigned_count": 26
    }
  },
  "metadata": {
    "source": "Companies House Public Data API",
    "licence": "Open Government Licence v3",
    "endpoint": "/company/00445790/officers",
    "attribution": "Contains public sector information licensed under the Open Government Licence v3.0. Source: Companies House."
  }
}

Every response includes a metadata envelope with source attribution, licence information, and retrieval timestamp.

Data source

All data comes from the Companies House API and is published under the Open Government Licence v3.0. Crown copyright.

Caching

Responses are cached locally in SQLite to reduce API calls and improve latency:

Endpoint	TTL
Company search	5 minutes
Company profile	1 hour
Officers	30 minutes
PSC	30 minutes
Filing history	30 minutes
Charges	30 minutes
Insolvency	30 minutes
Officer search	5 minutes
Officer appointments	30 minutes
Disqualified officer	1 hour
Advanced search	5 minutes
Registered address	1 hour

Cache is stored in ~/.bartram/cache.db by default. Set BARTRAM_DATA_DIR to change the location.

Development

git clone https://github.com/RobertAlsop/bartram_foundry.git
cd bartram_foundry
uv sync --all-extras
uv run pytest
uv run ruff check .

Licence

MIT

---

Built by Bartram Foundry — production-grade MCP tools for UK public data.