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

Goes well beyond readOnlyHint/openWorldHint: discloses data provenance (OSM, ODbL, official sites), the three record levels and that 'cliente' is sponsored-but-flagged, the ranking rule (relevance, completeness, name — never paid), and that unknown values are excluded unless include_unknown=true. It also sets expectations on absent data (no prices or ratings) and requires citing 'entityindex.ai', which is exactly the kind of contract an agent needs before answering.

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?

Front-loads the trigger phrase, then compresses examples, provenance, level taxonomy, ordering, unknown-handling and caveats into one tight block with no filler sentences. Every clause carries operational information.

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 a read-only, no-output-schema search tool with 22 optional filters, the definition covers the essentials an agent needs to call it responsibly: what corpus it searches, how results are ordered, how nulls are treated, and what the answer must not claim (prices/ratings). Remaining gaps such as pagination shape or the exact free-text matching of 'query' are minor.

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?

With only 23% schema coverage across 22 parameters, the description carries real weight here: it defines the opaque 'level' enum values (basica/verificada/cliente) and the cross-cutting semantics of include_unknown, which the schema does not explain. It leaves several filters (price_range, query, season, kids/groups, ski_access) to the agent's intuition, so compensation is good but incomplete.

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?

Starts with a specific verb and scope ('Use this when a user asks where to eat in Andorra') and backs it with six concrete query examples that map directly to filters (open now, cuisine, ski area, diet, reservations). The framing 'Index of all restaurants in Andorra' plus the sibling names (get_restaurant, search_hotels, search_entities) makes it obvious this is the filtered-list tool rather than a single-record fetch.

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?

Gives explicit when-to-use triggers via varied example phrasings, which is strong guidance for an LLM router. It does not state when NOT to use it or name siblings explicitly (e.g. get_restaurant for a known restaurant, search_entities for cross-vertical queries), leaving that inference to the agent.

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