patent_search

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

Consistent with readOnlyHint/openWorldHint and adds substantial context annotations cannot carry: per-backend auth requirements (EPO_OPS_CONSUMER_KEY, USPTO_API_KEY), the auto-mode concurrency and single-fallback-to-uspto_odp behavior, the fact that degradation surfaces in `errors`/`warnings` out of band, that every `patents` element is real and respects `limit`, and that seminal_prior_art is auto-mined from top-result co-citations. This is exactly the invisible 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?

The backend inventory is a single pipe-delimited run-on with stacked parentheticals that is hard to parse, and the quoting advice is duplicated verbatim from the `query` schema description. It is front-loaded correctly and most sentences carry information, but the density costs readability.

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?

Despite having 11 parameters and no output schema, the description covers the response shape (patents, errors, warnings, seminal_prior_art), credential prerequisites, fallback behavior, and latency tradeoffs. An agent has everything needed to call this correctly and interpret partial results.

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%, so baseline is 3; the description still adds real meaning beyond the schema by explaining what `source` selection implies (forced backend vs. auto) and that sort_by relevance ranking 'currently affects the ppubs backend only', which the enum alone does not convey. Filter parameters like cpc/assignee/date_range are left entirely to the schema.

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?

Opens with a specific verb+resource+scope: 'Search patents worldwide (US, EP, WO, JP, and 100+ authorities)', and it further carves out what kind of search it is (full-text conceptual, bibliographic, prior-art discovery). Against sibling patent_get the search-vs-fetch distinction is unambiguous, so an agent can route without opening the 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?

Gives clear operational guidance on backend choice: what each backend covers, what credentials it needs, what the default auto mode does, and 'Pass source to force a specific backend'. It also flags the seminal=true cost (~5-30s) and when to disable it. It stops short of naming an explicit alternative tool (e.g. patent_get for a known patent number), so it is not a full when/when-not treatment.

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