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

Annotations only declare readOnly=false, idempotent=true, destructive=false, openWorld=true; the description goes far beyond by disclosing processing semantics (leading covers unstamped, active content removed, text preserved, no OCR), refusal conditions (encrypted input, unreadable stamps), a 78 s execution bound, per-exhibit pricing, a no-logging stance with 24 h retention only for output=url, and an explicit prompt-injection warning about returned page text. That is precisely the kind of behavioral context annotations cannot carry.

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-loaded with purpose, then outputs, then inputs and limits, so the important routing information comes first. It is dense and telegraphic — several sentences are keyword fragments — which costs a little readability, but essentially every line (limits, refusals, pricing, retention) carries decision-relevant information rather than filler.

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?

Although an output schema exists and return values need not be described, the description still names the artifacts and the url-vs-base64 delivery tradeoff. Capacity limits, timeout, cost, refusal behavior, retention policy and an untrusted-content warning together leave no material gap for an agent considering an 8-parameter, nested-manifest tool.

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. The description adds only partial meaning: it labels manifest/numbering/covers_count as required and the rest as optional, and clarifies that leading covers are unstamped and 'caller chooses whether they count' (sharpening covers_count) plus that output=url yields 24-hour links. It does not explain numbering modes, zero_pad, or stamp_prefix composition beyond what the schema text already says.

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+resource pair ('produce stamped PDF exhibits and a checked index') and immediately enumerates the returned artifacts (exhibit PDFs, checked CSV/JSON indexes, optional bookmarked bundle). That combination of stamping plus index generation distinguishes it from the adjacent PDF utilities such as pdf_page_numbers, pdf_merge or bundle_pdf_citations without needing to open any 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 conveys usage context through hard constraints (1-20 PDFs, 500 physical pages, size ceilings, 78 s bound, refusal of encrypted input, no OCR) and a pricing model, which implies when the tool is a fit. It never names a sibling or states 'use this instead of X when Y', so an agent must infer routing from constraints alone.

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