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

Annotations already declare readOnly, idempotent, openWorld and non-destructive, so the safety bar is low, yet the description still adds real behavior: result cap of 20 with relevance ordering, snippet plus company apply link, a prompt-injection warning that titles/snippets are untrusted, pagination mechanics, and a data-retention disclosure. The one gap is that it doesn't say what an empty result set looks like, which matters for an openWorld search.

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?

Four sentences, front-loaded with what is searched and how results are returned, then pagination, then trust and privacy caveats. Every sentence carries information an agent would otherwise have to guess, though the logging/privacy sentence is peripheral to tool selection.

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?

With an output schema handling return shape, rich annotations covering the safety profile, and 100% schema coverage on all 7 parameters, the description supplies the remaining pieces: corpus scope, semantic matching, paging via next_offset, and untrusted-content handling. Nothing an agent needs in order to call this correctly or safely 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 baseline is 3, but the description adds meaning the schema lacks: the query is matched by meaning rather than keyword, so typos and related titles still hit, and it ties next_offset to offset to explain page chaining. The interaction between the documented default limit of 10 and the stated 20-result ceiling is left for the agent to reconcile.

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 a specific verb and resource ("Searches current job postings") and scopes the corpus to companies' own career sites and ATSs, plus semantic matching. It never names the siblings get_job or watch_search, so an agent must infer the search-vs-fetch-vs-save split from the names 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?

Usage is implied by the corpus description ("current job postings") and made concrete for pagination ("For more, pass next_offset back as offset"), which tells the agent how to continue iterating. There is no explicit when-not guidance or routing to get_job for a single posting or watch_search for recurring searches.

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