invowerk

Check an e-invoice

validate_invoice

Read-only

Check an e-invoice against EN 16931 and German rules.

Input: invoice_base64 is the base64 of an XML invoice (UBL or CII) or a ZUGFeRD / Factur-X PDF, at most 4 MB decoded; or pass the file as a download link in file. explain=true adds an explanation to findings whose rule has one, in lang ("de" or "en").

valid is the result. For XRechnung the KoSIT validator decides. If it is unreachable, invowerk's own check with the same rules decides, verdict_source is "local" and official_degraded is true. disagreement is true when both ran and differ. findings lists each problem with rule_id, severity and message; pass the codes to explain_errors (free). pdfa_verdict is the PDF/A check of a PDF. hints never change valid. Unreadable input returns a finding, not an error.

Credits: 1 per call.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`file`	No	The file as a download link, instead of `invoice_base64`. In ChatGPT, pass the user's uploaded file here. Other clients can set `download_url` to a public https URL.
`lang`	No	Language of the explanations: "de" (default) or "en".	de
`explain`	No	true adds an explanation to each finding whose rule has one.
`invoice_base64`	No	The invoice file, base64-encoded. At most 4 MB decoded.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed8 schema fields changedSep 28, 2026

addedInput schema / properties / explain / description

Added value: +"true adds an explanation to each finding whose rule has one."

addedInput schema / properties / file

Added value: +{
+  "default": null,
+  "description": "The file as a download link, instead of `invoice_base64`. In ChatGPT, pass the user's uploaded file here. Other clients can set `download_url` to a public https URL.",
+  "properties": {
+    "download_url": {
+      "description": "A public https URL of the file.",
+      "type": "string"
+    },
+    "file_id": {
+      "description": "An identifier for the file. ChatGPT sets it; other clients may pass any label, such as the file name.",
+      "type": "string"
+    },
+    "file_name": {
+      "type": "string"
+    },
+    "mime_type": {
+      "type": "string"
+    }
+  },
+  "required": [
+    "download_url",
+    "file_id"
+  ],
+  "type": "object"
+}

addedInput schema / properties / invoice_base64 / anyOf

Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]

addedInput schema / properties / invoice_base64 / default
```
Added value: +null
```

addedInput schema / properties / invoice_base64 / description

Added value: +"The invoice file, base64-encoded. At most 4 MB decoded."

removedInput schema / properties / invoice_base64 / type
```
Removed value: -"string"
```

addedInput schema / properties / lang / description

Added value: +"Language of the explanations: \"de\" (default) or \"en\"."

removedInput schema / required

Removed value: -[
-  "invoice_base64"
-]

First observedSep 28, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With only readOnlyHint/openWorldHint annotated, the description carries the real behavioral burden and delivers: KoSIT-vs-local verdict fallback, verdict_source, official_degraded, disagreement semantics, that hints never affect valid, that unreadable input yields a finding not an error, and a 1-credit cost.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose, then inputs, then result semantics, then cost. Dense but nearly every sentence carries operational information; only a couple of clauses (e.g. restating the 4 MB limit already in the schema) are mildly redundant.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a validator with an output schema, nested file object and external-validator complexity, the description covers input forms, degraded-mode behavior, disagreement handling and cost — nothing an agent needs to call it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is already 100%, so the baseline is 3, but the description adds genuine meaning: the 4 MB decoded cap, that explain only annotates findings whose rule has one, and that lang de/en controls explanation language, plus the either-or relationship between file and invoice_base64.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (check/validate), resource (e-invoice) and the exact standards applied (EN 16931, German rules). This cleanly separates it from siblings like parse_invoice, convert_invoice and render_invoice without ambiguity.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explains the two mutually exclusive input modes (base64 vs. download link) and points to explain_errors for free code explanations. It lacks an explicit 'use this instead of parse_invoice when...' routing statement, so it stops short of full when-not guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All invowerk tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.