Start a crawl

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

With annotations already declaring it non-read-only, open-world, non-idempotent, the description adds real behavioral context beyond them: the async taskId return, that the 100-page default is echoed back, that retention duration is unknown to the client, and that collapsed/refused links are still tallied in get_crawl's discovery. It does not discuss idempotencyKey semantics or rate/robots constraints.

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?

Five dense sentences with the traversal scope front-loaded and the deferred workflow immediately after; little padding. The 'Returns { taskId }' sentence slightly overlaps an existing output schema, but the appended maxPages detail makes it earn its place.

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 40-parameter tool with annotations and an output schema, it covers the essentials: what it traverses, what it returns immediately, how to retrieve results, and where page-option semantics live. The remaining gaps (most boolean/array flags, retention policy) are acknowledged or handled by the schema rather than silently omitted.

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 40 parameters at 50% schema coverage, the description can only partially compensate. It adds meaning for maxPages (the default comes back in the response), maxDepth, and URL folding, and cross-references scrape for the page options, but the bulk of the option surface is left 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?

It names a specific verb and resource ('Start crawling a section of one site') and bounds the scope precisely (start path's subtree, own host + www twin). That scope description implicitly separates it from scrape (one page) and map (URL discovery only), though it never names those siblings explicitly.

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 clear operational context: it returns { taskId } at once, so wait_crawl, then get_crawl_pages, is the intended follow-up chain. It also routes page-option questions to scrape's parameter descriptions. It stops short of stating when NOT to crawl (e.g. use map for URL listing, scrape for a single page).

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