scrape_url

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

Annotations already declare readOnlyHint/openWorldHint/idempotentHint and no destructiveness, so safety is largely covered by structured data. The description does add genuinely non-obvious behavior: ~120s scrape wait, per-mode cost (~1 credit vs ~60 credits/page with proxy), credit consumption in the workspace, and the requirement to pass both useProxy and confirmExpensiveProxy. That extra cost/limit context is valuable and, combined with a true read-only profile, warrants a 3 without over-crediting repetition of the annotations.

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 default mode and its return shape, then the scrape mode and the post-scrape workflow, then proxy rules. The proxy warning is repeated two or three times (useProxy paragraph, CRITICAL paragraph, plus the schema), which is justified by the cost risk but slightly inflates length.

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?

Given five parameters, no output schema, and a potentially expensive procedure, the description covers the return format, timing, cost model, workspace resolution ('never pass workspaceId'), and downstream steps (check image URLs then call read_image). 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%, so the baseline is 3, but the description adds meaning beyond the schema: it frames mode=check as the default, bounds the batch at 20 URLs for check mode, explains the single-URL restriction for scrape mode, and ties useProxy to both an explicit user confirmation and the companion confirmExpensiveProxy flag. That conjunction constraint is the kind of cross-parameter relationship that schemas alone rarely convey.

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?

States specific verbs and resources (validate/ping URLs vs full page scrape) and cleanly separates the two modes by outcome: status/content-type/final URL versus text/colours/favicon/images. An agent can distinguish it from siblings like create_kb_from_urls or read_image using this text alone.

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 guidance for each mode (verify widget images, logos, CDN assets, booking links; scrape when building branded agents) and explicitly names alternatives/exclusions: 'Do NOT use scrape for KB ingest (use create_kb_from_urls)' and 'call read_image on every important photo'. It also states the proxy preconditions and forbids proactive use.

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