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

Annotations are sparse (non-read-only, non-destructive, open-world false), so the description carries most of the burden and delivers richly. It surfaces silent clamping behavior for several fields, tier-gated 402 behavior for AVIF/urls mode, accepted-but-unused watermark parameters, non-1-bit-JPEG fallback behavior, and special default interactions (e.g., dpi=150). This goes well beyond what annotations or schema types convey and materially prevents incorrect agent expectations.

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 effectively one long sentence with a very long serial list of features. It front-loads the core purpose well, but the tail becomes dense and list-like, making the description harder to parse quickly. Every clause carries information, yet the structure does little to group related concepts beyond the '[category: pdf]' tag. It is not bloated or redundant, but it is at the edge of what counts as concise.

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?

For a 30-parameter tool with no output schema, the description covers the main capability areas, tier constraints, output formats, and response modes, and the schema covers every parameter with detailed behavioral notes. The description plus schema fully equips an agent to invoke this tool, including knowing which parameters are gated by tier. The only mild gap is the lack of a mention of the return format—whether the output is a download link, base64 payload, or something else—but the response-mode parameter (zip/json/urls) largely addresses that.

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 baseline is 3. The description adds meaningful cross-parameter context not present in the schema: it tells the agent that preset profiles bundle multiple settings and that explicit fields override presets, and it highlights mode/sheet/filmstrip relationships. It also mentions response modes and filename patterns at a higher level, reinforcing usage intent. It does not restate every parameter, which is appropriate because the schema already documents them.

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 specific verb-resource pair ('Rasterize PDF pages to PNG / JPG / WEBP / TIFF') and uses a title-like lead-in that names the tool's domain. It enumerates a large set of feature areas (page ranges, DPI, color modes, tile mode, presets, response modes), which makes the tool's exact scope clear and distinguishes it from nearby PDF-related siblings like pdf_thumbnails, pdf_extract_pages, or convert_* tools. The '[category: pdf]' tag reinforces grouping without substituting for meaning.

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 makes the primary use case obvious (rasterize PDF pages into image formats) and includes enough feature detail to signal when it applies, but it does not explicitly say 'use X instead for Y' or state when not to use it. With roughly 130 siblings, the absence of named alternatives is a gap, but the detailed feature list still gives an agent solid contextual guidance. Tier-limited options (Starter+) are called out, which helps with routing when plan constraints matter.

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