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

Annotations cover the safety profile (readOnly, idempotent, non-destructive, non-openWorld), and the description adds context they cannot: data provenance ('the authors' own views and accounts, never GPS findings'), a mandatory attribution obligation, PII redaction ('names of people in custody and GDC numbers are removed before the data leaves GPS'), and the semantics of the `counts` output. These are real behavioral traits beyond the structured fields.

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?

Purpose and source are front-loaded, and the filter list with defaults reads as a compact reference block. The four label definitions are necessary but make the prose dense and slightly sprawling; a bulleted form would scan faster, though nothing is truly wasted.

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 no output schema, the description compensates by enumerating returned fields (author, date, link, label) and explaining the `counts` aggregate. Combined with full filter semantics, ordering, provenance and privacy handling, an agent has everything needed to call and interpret the tool correctly.

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 0%, so the description carries the full burden and does so: it defines q, facility (with an example, 'Telfair'), days (default 30, max 365), label (enumerating account | opinion | idea | news) and limit (default 10, max 60). Every parameter's meaning, default and bound is supplied where the schema gives nothing.

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 (public Facebook posts and comments) and pins down the exact data source: the Ga. Prisons Exposed group plus GPS's own posts. The label taxonomy and the 'newest first' ordering let an agent distinguish this from siblings like search_quotes or search_site 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 framing ('What people on Facebook are saying about Georgia's prisons') and the full filter list give a clear sense of when this tool applies, and the attribution note ('attribute and link every quote') constrains use. However, no alternative tool is named and there is no explicit when-not guidance despite 22 siblings.

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