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

Annotations declare readOnlyHint=false, openWorldHint=true, idempotentHint=false, and destructiveHint=false, but the description goes far beyond by disclosing the asynchronous polling model (returns immediately, 1-2 minutes until done, poll get_advanced_seo_report), the refusal condition (already running), the cost (1 AI brain credit, charged upfront, refunded on provider failure), and subscription requirement. It also describes the side effect of storing a report, which is richer than the bare 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?

The description is dense but every sentence earns its place: it starts with the core action, then covers async behavior, prerequisites, refusal, cost, subscription, and parameter guidance. There is no filler, and the structure flows logically from what → how → when → prerequisites → cost. Despite its length, it remains highly scannable and front-loaded with the primary purpose.

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 the tool's asynchronous nature, cost implications, required keyword precondition, and interplay with sibling tools, the description covers all necessary details: what the analysis produces, how to retrieve results (polling), when to rerun, and account context (website_id). While an output schema exists, the description also enumerates the report fields, ensuring an agent understands the outcome without needing to open the schema.

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?

The schema already provides 100% coverage: both parameters have descriptive text ('Article id from list_articles.' and 'Website id from get_account. Optional when the account has a single website.'). The description repeats the website_id guidance ('Pass website_id when the account has several websites'), which adds marginal value beyond the schema. Since the schema carries the meaning, the baseline of 3 is appropriate; the reinforcement does not elevate it.

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 opens with a precise verb ('Starts') and resource ('competitor-relative SEO analysis of the default-language article'), then specifies the exact steps (fetches top Google results, compares depth and topic coverage, stores a report). This clearly differentiates it from siblings like check_article_seo_score (which reads the report) and improve_article_seo (which acts on it), and the distinction is reinforced by the note that these tools use the benchmark automatically.

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 provides explicit when-to-use guidance: the article must have a target keyword, it is refused while an analysis is already running, and it should be re-run only when the article changed substantially (with a pointer to is_stale). It also instructs to pass website_id when the account has several websites (linking to get_account). While it does not name a direct alternative tool, it clarifies that check_article_seo_score and improve_article_seo consume the benchmark, so rerunning is unnecessary unless content changed.

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