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

Annotations already declare readOnly/idempotent/non-destructive/open-world, but the description adds substantial context they cannot: a precise cost model (1 credit per 10 returned URLs, 1 credit per page read, failed reads not charged), graceful degradation ('unreachable pages do not break the crawl'), a 20 000-char per-page cap, and a skipped-with-reason list. This is exactly the beyond-annotation disclosure expected.

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 content is well front-loaded (purpose, when/when-not, filters, returns, cost), but every sentence is duplicated in Russian and English, roughly doubling the length for no informational gain. The Russian block also adds a subtle semantic nuance about search vs selectPaths/URL filtering not fully mirrored in English, so the redundancy is not purely wasteful but still heavy.

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 6-parameter open-world reader with no output schema, the description supplies everything an agent needs: return shape (pages URL/title/text plus skipped with reasons), failure behavior, cost, and the interaction between the text filter and URL filters. Nothing material is left unstated.

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 the baseline is 3, but the description earns more by clarifying semantics the schema only partially conveys: search matches page TEXT/title rather than URL, selectPaths/excludePaths match URL substrings, and the warning that omitting selectPaths makes crawl read the map's first (often service) URLs. That is meaningful disambiguation beyond the field descriptions.

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 opening states a specific verb and resource — crawl reads a site's URL map then the page text — and frames it as one call replacing the manual find-link/read loop. It explicitly distinguishes itself from siblings map and read_url, so an agent can select it 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?

There are dedicated 'Use when' and 'Do not use when' sections naming concrete triggers ('collect every event from this aggregator', 'read this docs section') and the alternatives that win in the excluded cases (map for URL lists, read_url for one page). It even prescribes the recommended call order (map first, then crawl with selectPaths/search).

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