Check for broken links

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

Annotations already declare readOnlyHint=true and destructiveHint=false, so the agent knows this is a safe read operation. The description goes well beyond annotations by revealing several non-obvious behavioral traits: internal-only default with opt-in external checks, automatic skipping of private/loopback/link-local hosts, bounded concurrency, per-call link cap, the cost implication ('each checked link costs 1–2 outbound requests, which matters on Cloudflare Workers'), and the cursor-based continuation mechanism. These are exactly the kind of behavioral details an agent cannot infer from annotations or schema alone.

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 description is dense but efficiently organized: it front-loads the core function ('Find broken links and images in rendered content'), lists the failure categories, states the three scan modes, clarifies the default scope and opt-in external behavior, and ends with the crucial pagination/concurrency caveat. Every sentence earns its place. No filler or tautology. The structure moves from 'what' to 'how' to 'when to know you're not done', which is ideal for an agent deciding whether and how to call this tool.

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 12 parameters and no output schema, the description answers the key questions an agent would have: what to pass to scope the scan, how pagination works, what the cost/limit behavior is, and what safety restrictions apply. The absence of an output schema means the description doesn't need to explain return values, but the description still tells the agent enough about the response shape ('follow next_cursor for the rest') for correct usage. It also mentions the `next_cursor` field by name, signaling what the response contains. This is complete for a read-only scanning tool of this complexity.

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 description coverage is 100%, so the schema already documents every parameter's meaning. The description adds context around the parameters (e.g., that 'a content type' corresponds to the `type` param and that 'one front-end URL' corresponds to `url`), and it explains the relationship between `ids`, `type`, `url`, `limit`, `status`, and `cursor`. However, most of the per-parameter semantic weight is already in the schema, so the description's added value is contextual rather than definitional. Baseline 3 is appropriate for full schema coverage; the description provides some context but doesn't add deep parameter nuances beyond 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?

The description clearly identifies the tool's purpose: finding broken links and images in rendered content, and enumerates specific categories of failures (4xx/5xx, network failures, redirect chains, mixed content, links to draft/trashed/missing content). It also specifies scan targets ('given ids, a content type, or one front-end URL'), which distinguishes it from sibling tools like check_accessibility or internal_link_report. It uses a precise verb ('find broken links and images') and resource ('rendered content'), so an agent can tell it apart 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?

The description explicitly states what it scans ('given ids, a content type, or one front-end URL'), explains the default scope ('internal links only by default; external checks are opt-in'), and flags the important constraint that external checks 'never reach private or loopback addresses'. It also implies when to use it versus alternatives like check_accessibility, internal_link_report, or seo_site_check, all of which are sibling tools with related but different functions. The 'Scans given ids, a content type, or one front-end URL' sentence serves as a clear usage condition. It also explains the pagination pattern ('follow next_cursor for the rest'), which is essential for correct use on large scans.

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