Skip to main content
Glama

Server Details

Read incoming supplier invoices through a remote MCP server and get structured data for accounting.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
Last Tested
Transport
Streamable HTTP · MCP 2025-11-25
URL
Repository
ANTONDrakon/duebuddy-mcp
GitHub Stars
0
ChatGPT
–
Claude
–

TDQS

A4.6/5.0

Scored across 4 tools

Disambiguation5/5

Each tool has a clearly distinct job: parse a document, poll a job, check the prepaid counter, and fetch a top-up page. The explicit 'Use when / Do not use when' pairs (e.g. get_result vs parse_invoice, get_balance vs topup_link) cleanly resolve the only near-overlaps.

Naming Consistency4/5

get_balance, get_result and parse_invoice follow a clean verb_noun pattern, and all names are snake_case. topup_link is the lone deviation, naming a resource rather than an action (get_topup_link/create_topup_link would be more parallel), but it stays readable.

Tool Count5/5

Four tools exactly cover the workflow loop: check balance, parse, poll result, top up when empty. Nothing is redundant, and no helper tool is missing for the stated scope.

Completeness4/5

The full read-and-bill lifecycle is covered, including status polling, export links, throttling, and balance recovery. Minor gaps exist: no way to list prior jobs/documents or cancel a queued job, so an agent that loses a job_id has no recovery path.

Available Tools

4 tools
get_balanceCheck prepaid balanceA
Read-onlyIdempotent
Inspect

Shows how many scanned pages are left on the prepaid balance for this API key.

Use when: before a batch of scans or photos, or when parse_invoice answered insufficient_balance and the remaining amount must be confirmed. Do not use when: the user asks about invoice content — that is get_result.

Side effects and cost: none. The number comes from the same counter that blocks parsing, so it can change between two calls while a batch is running.

Parameter interactions and edge cases: this is the same counter that blocks parsing, so a balance of zero makes the next parse_invoice fail with insufficient_balance and a topup_link is the only way forward. Two keys of one account see the same number, because usage is counted per account. The number can drop between two calls while a batch runs, and pages parsed on the free cabinet allowance are not part of it.

Returns: {ok, balance} with remaining pages and the current plan.

Inputs: api_key.

get_resultGet parsed invoice resultA
Read-onlyIdempotent
Inspect

Returns the result of a parse job started by parse_invoice: structured fields (supplier, invoice number, dates, net/VAT/gross, currency, line items) plus export links for accounting systems.

Use when: you already have a job_id and want its status or its extracted data. Do not use when: you have no job_id yet — start with parse_invoice. Do not poll in a tight loop: one call every few seconds is enough, text PDFs and XML usually finish in under a minute.

Side effects and cost: none. Does not touch the prepaid balance, so repeated polling is free; a failed job stays failed.

Parameter interactions and edge cases: job_id belongs to the key that created it, so a job started with one key is invisible to another. A pasted URL or a document id from another system returns 404 rather than a guess. A job that failed stays failed — polling it again will not resurrect it; start a new parse_invoice call instead.

Returns: {ok, result}, where result.status is queued, processing, done or failed; on done it also carries fields and export links.

Inputs: job_id (required), api_key.

parse_invoiceParse one incoming invoiceAInspect

Reads ONE incoming supplier invoice and returns a job id for polling.

Use when: the user has a single invoice (PDF, photo, scan, XML e-invoice or QR bill) and wants supplier, invoice number, dates, amounts, VAT or line items extracted. Do not use when: the user asks to ISSUE an invoice, to send a payment, or to post entries into DATEV/1C — this tool only reads documents. For several documents call it once per document; the returned job ids can be polled independently.

Side effects and cost: the document is downloaded and queued. XML and QR invoices are parsed by code and cost nothing; scans and photos cost 0.07 EUR per page, charged to the prepaid balance. Not idempotent: each call creates a new job and a new charge for scans.

Parameter interactions and edge cases: the same file parsed twice returns the existing job with charged=false instead of billing twice, so retrying after a client timeout is safe — and the second call's country is ignored, the first profile wins. Per API key there is a request rate limit and a cap of 3 parses in flight; exceeding either returns 429 with Retry-After, which is a throttle, not a billing error. country changes field names and tax checks, never the price.

Returns: {ok, job_id, status}. Next step: call get_result with that job_id.

Inputs: file_url (required), api_key, country.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables converting invoice PDFs, scans, and photos into Excel/CSV spreadsheets, checking invoice arithmetic, extracting structured JSON data, and retrieving outputs from previously submitted documents through a hosted MCP service.
      297 npm
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      MCP server for e-invoice platforms. Enables natural language querying of invoices, partners, company data, and financial reports.
      18
      11 npm
      6
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Queries official Brazilian federal received electronic invoices (NFS-e) via a hosted, read-only MCP server with prepaid credits, requiring no platform credentials.
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Remote MCP server for extracting structured data from images, PDFs, and XML documents using LLM Vision models. Provides specialized tools for demographic and invoice data extraction with support for multi-part documents.
      -
    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.