Bestand zählen und summieren

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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds operational context that annotations cannot: a cap of 10 API calls, that list_direction 'both' doubles them, that the token bucket is process-local and not shared between two clients on the same tenant, and that sums/groups only appear when the set was read completely. The payment_status caveat (amount_paid measured as 0.00 even for paid receipts, partial payments invisible) is exactly the kind of hidden behavior an agent needs.

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

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

Purpose is front-loaded and every sentence carries weight — replacement scope, caveats, limits, token-bucket warning. The sentences are long and semicolon-chained, which costs some scannability, but there is no filler.

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

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

For an 11-parameter aggregation tool with an output schema, the description covers the remaining risk surface: which sibling it supersedes, the API-call budget, the 'both' expansion, the completeness precondition for sums, and data-quality traps. Nothing needed to invoke it correctly is missing.

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

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

Schema coverage is 100% and the schema itself is unusually detailed, but the description still adds meaning beyond it: it discloses that 'both' is not an API value but a server-side construct rendered as two separate runs (relevant to the call budget), and it frames max_rows as a presentation knob while all pages are still read server-side. These are behavioral semantics the schema does not carry.

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

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

The first clause names the specific verb+resource ('zählt serverseitig über alle Seiten von Belegen, Zahlungen oder Buchungen') and states the distinctive output contract (Anzahl und Summe statt aller Zeilen). It then explicitly names the three sibling tools it replaces, so an agent can separate it from bb_receipts_search, bb_transactions_search and bb_postings_search without opening any schema.

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

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

It gives concrete questions the tool answers ('wie viele offenen Eingangsrechnungen gibt es', 'was ist im März über PayPal gelaufen') and draws an explicit boundary: this tool for counting/summing, the *_search tools remain responsible for single fields, sorting and further filters. When-to-use and when-not-to-use are both present.

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