mcpField
L
ledgerbackup

ledgerbackup

Local
github.com/CloudCorpRecordsFinance0Updated 1w ago

About ledgerbackup

ledgerbackup is a local bridge for users who need read-only exports of accounting data from Boxkite-connected organisations. It forwards tools to Boxkite's hosted MCP server and tracks exports through their manifests and download URLs.

Key features

  1. 01List connected accounting organisations
  2. 02Start read-only organisation exports
  3. 03Track export status and progress
  4. 04Browse manifests and file mappings
  5. 05Retrieve the finished ZIP download URL

Use cases

  1. 01Check which organisations are available
  2. 02Request an organisation data backup
  3. 03Inspect files included in an export
  4. 04Download a completed export

MCP support

Exposes the upstream Boxkite tool list locally and forwards MCP tool calls to its hosted server, passing successful results through unchanged.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"cloudcorprecords-ledgerbackup": {
"command": "npx",
"args": [
"-y",
"@reneturcios/ledgerbackup"
]
}
}
}
Run commandnpx -y @reneturcios/ledgerbackup

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@reneturcios/ledgerbackup0.1.0
ocighcr.io/cloudcorprecords/ledgerbackup:0.1.0

Tools

7

Saved tool list · updated 1w ago

This tool takes no parameters.

Details

From the project README

ledgerbackup

ledgerbackup is a small local bridge to the hosted Boxkite MCP server, which backs up and exports accounting data from the organisations connected to your Boxkite account (Xero, Sage, Reckon, or QuickBooks). It lists those organisations, starts a read-only export of one organisation's data, and then tracks that export through its status, manifest and file-to-object mapping until a signed download URL for the finished ZIP is available.

The bridge keeps no accounting data of its own. It serves the upstream server's tool list locally, exactly as captured from the live endpoint, and forwards each tool call upstream, returning the result untouched.

Tools

  • list_organisations — list the accounting organisations connected to the authenticated Boxkite account. No arguments.
  • start_export — start a read-only export of one organisation's data. Required: organisationId (from list_organisations). Per the upstream description, an export costs $39 per organisation: when payment is outstanding the response is status awaiting_payment with a Stripe checkout URL, and repeat calls return the existing export rather than creating or charging for a second one. A typical export runs for hours.
  • list_exports — list one-off exports, newest first, optionally filtered by organisationId. Side-effect free: it never starts an export or charges anything.
  • get_export_status — report one export's state (queued | running | ready | failed), its progress, and when to poll again. Required: exportId.
  • get_export_manifest — see what an in-progress export contains so far, paged via fileOffset and fileLimit. Required: exportId.
  • get_export_mapping — for an in-progress export, map each discovered file to the accounting object it represents, paged the same way. Required: exportId.
  • get_download_url — return a short-lived signed URL for the finished export ZIP, valid within the export's 7-day retention window. Required: exportId.

Configuration

  • BOXKITE_MCP_URL (optional) — override the upstream endpoint. Defaults to https://app.getboxkite.com/mcp. (The override hook is this rebuild's own labelled reconstruction.)
  • BOXKITE_API_KEY (optional) — when set, every upstream request carries an X-API-Key: <key> header with your Boxkite API key. When it is not set, no key header is sent. No Authorization header is ever sent. (This forwarding is also a labelled reconstruction; the original surface was captured keyless.)
Forwarding mechanics

On the first tool call the client opens a streamable HTTP session with the upstream: a JSON-RPC initialize request (protocolVersion 2025-03-26, clientInfo name ledgerbackup), then a notifications/initialized notification, then the tools/call request carrying the caller's arguments unchanged. When the upstream answers initialize with an Mcp-Session-Id header, that value rides on every later request of the session, which is reused for the life of the process. Replies may arrive as plain JSON or as server-sent events; both are read.

Successful upstream results pass through verbatim, including any isError flag the upstream sets. Failures — a JSON-RPC error, a non-2xx status, an unreadable response, or a request that cannot be completed — come back as an isError result whose text starts with ledgerbackup error: (this rendering is this rebuild's own labelled reconstruction).

CLI

Each tool is also a subcommand taking one JSON argument string:

npx @reneturcios/ledgerbackup list_organisations
npx @reneturcios/ledgerbackup start_export '{"organisationId":"org-123"}'
npx @reneturcios/ledgerbackup get_export_status '{"exportId":"exp-9"}'
npx @reneturcios/ledgerbackup get_download_url '{"exportId":"exp-9"}'
npx @reneturcios/ledgerbackup --version

Exit codes: 0 on success, 1 on an isError result, 2 on a usage error.

MCP server

npx -y @reneturcios/ledgerbackup mcp

MCP client config:

{
  "mcpServers": {
    "ledgerbackup": {
      "command": "npx",
      "args": ["-y", "@reneturcios/ledgerbackup", "mcp"]
    }
  }
}

Docker

docker run -i ghcr.io/cloudcorprecords/ledgerbackup
docker run --rm ghcr.io/cloudcorprecords/ledgerbackup --version

The container's default command is mcp, so the first form is a stdio MCP server; any other arguments reach the CLI normally.

Development

npm install
npm run build
npm test
npm run dev

The test suite runs entirely against a stubbed JSON-RPC upstream — no live network calls.

License

MIT — see LICENSE.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.