Skip to main content
Glama

ParliamentOS – Bundestag, Landtage, Nationalrat, EU

Belege prüfen

pruefe_belege

Prüft einen fertigen Text mit Beleg-Kennungen [[beleg:…]] gegen die amtlichen Zeilen (Zitat-Validator nach Beleg-Vertrag v11, deterministisch, ohne Sprachmodell) und stellt einen signierten Prüfschein mit eigenem Link aus. Gibt je Belegstelle das Urteil zurück (verifiziert, abweichend, nicht_gefunden, nicht_pruefbar), bei Abweichungen Merkmal, Sollwert laut Zeile und Angabe im Text. Aufrufen, wenn eine Antwort mit Belegen fertig ist. Bei abweichend die Stelle mit get_beleg nachlesen, korrigieren und den ganzen Text erneut prüfen; nicht verifizierbare Stellen als ungeprüft kennzeichnen oder streichen. Die Fußzeile aus der Antwort unverändert unter den Text setzen. Der Satz zur maschinellen Prüfung gehört nur unter einen Text mit Ergebnis freigegeben. Jeder Aufruf stellt einen Prüfschein aus; Konten im kostenlosen Tarif haben 20 Prüfscheine im Kalendermonat, das Prüfpaket 200. Die Antwort nennt den Stand oder, für Zugänge ohne Monatskontingent (API-Schlüssel, Professional, Enterprise, Organisationen), dass keines gilt. Sind sie aufgebraucht, gilt der Text als ungeprüft. Der Text selbst wird nicht gespeichert, nur sein Hash, die Kennungen und die Urteile. [Ohne KI: liefert ausschließlich Originaldaten aus amtlichen Quellen.]

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "text": "<text>"
    },
    "name": "pruefe_belege"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYesDer vollständige Antworttext mit allen Beleg-Kennungen [[beleg:…]], genau so, wie er dem Leser gezeigt wird. Höchstens 100.000 Zeichen.
regelnNoOptional: Kennung der Belegregeln, nach denen der Text entstand, z. B. skill-belegen@1.4.0 oder prompt-belegte_antwort. Wird nur gezählt.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.6/5.0
Behavior5/5

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

Far beyond the annotations: it discloses that every call issues a signed certificate (consistent with readOnlyHint=false and idempotentHint=false), states monthly quotas (20 free / 200 Prüfpaket) and the unlimited case for API keys, Professional, Enterprise and organizations, and defines what happens when quota is exhausted (text counts as ungeprüft). It also discloses data retention — only the hash, citation IDs and verdicts are stored, not the text.

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?

Purpose, return shape, remediation loop and quota are front-loaded in that order and every sentence carries operational information. It is dense and somewhat long, with the quota paragraph the only part that could be tightened, but nothing is filler.

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?

Despite having no output schema, the description explains the return shape (per-citation verdict, Merkmal/Sollwert/Angabe bei Abweichung, certificate link, quota status) well enough to consume the result. Side effects, retention and quota failure mode are all covered, so nothing an agent needs to call this correctly is missing.

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

Parameters3/5

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

Schema description coverage is 100%, so the input schema already defines both 'text' and 'regeln' including the character limit and the rule-ID examples. The description reinforces that the text must be the finished answer with citation markers, but never mentions the optional 'regeln' parameter, so it adds little meaning beyond the schema. Baseline 3 applies.

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?

The description names a specific verb and resource ('Prüft einen fertigen Text mit Beleg-Kennungen … gegen die amtlichen Zeilen'), identifies the mechanism (deterministic citation validator, no language model) and the output artifact (signierter Prüfschein mit eigenem Link). It also distinguishes itself from the read-sibling by prescribing get_beleg for the follow-up lookup, so an agent can tell the two apart without opening either schema.

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

Usage Guidelines5/5

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

The trigger is explicit ('Aufrufen, wenn eine Antwort mit Belegen fertig ist') and the follow-up branch is spelled out per verdict: on 'abweichend' re-read with get_beleg, correct, re-run the whole text; on non-verifiable spots mark as ungeprüft or delete. It also states how to place the footer and when the machine-check sentence may be used, which is exactly the when/when-not guidance an agent needs.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.